robinv
Joined: 31/08/04
Posts: 277
|
Vista performance testing
#433364 - 13/03/07 03:34 PM
|
|
|
With reference to an earlier thread about Vista (should we shouldn't we, will she won't
she etc) I said i'd post my performance findings once i'd given it a bloody good
thrashing.
Here they are:
http://rainrecording.co.uk/vista/performance
Yes, i do work
for Rain Recording, no, it's not just some marketing nonsence, yes we still do XP systems.
I hope you find this information useful - anyone else got any Vista
experiences to share?
Thanks
Robin
--------------------
http://www.pc-music.com
|
pocketrocket
Joined: 09/03/07
Posts: 8
|
Re: Vista performance testing
[Re: robinv]
#433548 - 13/03/07 10:06 PM
|
|
|
Very intesting. I'm more tempted to stick another drive in and dual boot 
What was the cpu usage like Xp vs Vista ?
|
Neil C
active member
Joined: 01/04/03
Posts: 2137
Loc: Designated cuddle zone
|
Re: Vista performance testing
[Re: robinv]
#433556 - 13/03/07 10:20 PM
|
|
|
That is fascinating.
|
table for two
active member
Joined: 24/03/02
Posts: 5808
|
Re: Vista performance testing
[Re: robinv]
#433570 - 13/03/07 10:45 PM
|
|
|
Thanks for that Robin
With the 8GB RAM,
32bit vista can utilise 4GB of
it, 64bit Vista 8GB.
Thus for instance the HalionOne sample streaming adavantage of
Vista vs XP is understandible.
Interesting that the track counts are the same
for XP & Vista
It would be nice to see, a say 2GB RAM & possibly a 1GB RAM
comparison, as many users have this config.
Also a C2D, not a quad, as many
users have a have a C2D, not a quad.
Also would be nice to see a Sonar
test.
e.g. Sonar 3 on XP with 2GB RAM vs Sonar 6 on Vista & XP with 2GB RAM.
As regards 32bit vs 64bit Vista :
to see the gains of 64bit Vista over
32bit, one would need the audio sw to be compiled for 64bit.
I suspect Sonar
6.x would shine on 64bit Vista.
Thanks again for the work Robin &
for posting it up.
|
robinv
Joined: 31/08/04
Posts: 277
|
Re: Vista performance testing
[Re: table for two]
#433587 - 13/03/07 11:12 PM
|
|
|
Quote table for two:
Thanks for
that Robin
With the 8GB RAM,
32bit vista can utilise 4GB of it, 64bit
Vista 8GB.
Thus for instance the HalionOne sample streaming adavantage of Vista vs
XP is understandible.
Actually with earlier testing on XP with 2GB, 4Gb and 8GB i saw very little difference
in performance in these tests. I'm on the lookout for something that relies heavily on
memory - i was hoping Gvi might do it but it wont install on Vista.
Quote:
Also a C2D,
not a quad, as many users have a have a C2D, not a quad.
Sure, but there are only so many hours in the day
and i wanted to go with the baddest machine i could muster
Quote:
Also would be nice to see a Sonar test.
e.g. Sonar 3 on XP with 2GB RAM vs
Sonar 6 on Vista & XP with 2GB RAM.
Yep, good suggestions, hope to do all that in time
Quote:
As regards 32bit
vs 64bit Vista :
to see the gains of 64bit Vista over 32bit, one would need the audio
sw to be compiled for 64bit.
I suspect Sonar 6.x would shine on 64bit Vista.
I did some preliminary testing
with Sonar 6.2 on Vista 64 and it seemed to only match XP32 - needs further investigation
though. Vista 64 "Compatible" and "Optimised" are not the same thing i guess.
Cheers
Robin
--------------------
http://www.pc-music.com
|
table for two
active member
Joined: 24/03/02
Posts: 5808
|
Re: Vista performance testing
[Re: robinv]
#433627 - 14/03/07 07:29 AM
|
|
|
For such tests, it is rather useful to test on what the average user might have.
Not
the more maxed Quad core, 8GB RAM as on the test.
And not just on C2D but also
Athlon 64 X2 AM2.
A laptop test is very useful. Comparing Centrino, Turion,
Turion X2, CD/C2D.
Undoubtedly Vista has been coded to take advantage of Dual
Core technology, associated changes in motherboard chipsets, PCI-E, SATA etc,
thus
it will have an adavantage over XP,
but this is Not the current issue for
musicians.
Stable Vista drivers for their audio cards is the issue.
As
is all their current audio software running fine on Vista.
And an overhaul of their
current stable working practices.
Further, the price of 2GB RAM sticks are
high.
Once they drop to below £100 then Vista becomes more feasible.
Looking before leaping with Vista is prudent imo.
|
young david
Joined: 25/11/06
Posts: 253
Loc: Stockholm
|
Re: Vista performance testing
[Re: robinv]
#434009 - 14/03/07 11:14 PM
|
|
|
Thanks for posting this Robin - very interesting.
I saw there's a nice table on
Rain's site under 'Vista watch' showing compatibility of various hardware. Seems to be
quite a bottleneck.
So Cubase 4 is working ok with both 32 and 64 bit Vista,
but what's the situation roughly speaking with 3rd party plugins?
Dave
--------------------
http://youngdavid.net
myspace
|
TAFKAT
member
Joined: 08/01/03
Posts: 130
Loc: Australia
|
Re: Vista performance testing
[Re: robinv]
#434051 - 15/03/07 03:27 AM
|
|
|
Quote robinv:
I hope you find
this information useful - anyone else got any Vista experiences to share?
Hey Robin,
Thanks for sharing
Mate.. :-)
Have you tested Nuendo 3 / SX 3 at all on Vista , it would really
be interesting whether the same performance deltas are maintained between XP32- Vista
32/64
I will diving into my Vista development /testing shortly with numerous
audio apps/hardware ,and I'll be more than happy to compare notes.
I'll be
using the existing Blofelds DSP40 and L- Factor test respectively, purely due to the
existing collated data base.
Your quite welcome to run the benches up yourself,
and see if the performance improvements that you experienced in C4, also translates to
N3/SX3, or whether Steini have added some extra hooks into C4 Vista optimisations.
Benches can be found Here
Peace:
V:
|
robinv
Joined: 31/08/04
Posts: 277
|
Re: Vista performance testing
[Re: TAFKAT]
#434184 - 15/03/07 11:59 AM
|
|
|
Thanks for the benchmark links - i'll check those out. They do seem to be down the same
lines as what i was trying to do. Measuring CPU usage, like with the five towers test,
seems to be a bit unreliable (or at least i think so) whereas working to achieve a maximum
stable playback gives, i feel, a better sence of what's possible on a given system and
also gives more interesting results. On some other forums my results have taken some
crticism but i think it may be down to people thinking in benchmarks rather than glitch
free playback. What i mean is that doing traditional benchmarking on the system might have
shown much closer performance results, but in reality the XP install was not as good at
maintaining a stable CPU usage of 4 cores in Cubase and so glitched much earlier than the
Vista install - you'd only know that if you were listening to it.
Much as i'd
like to test every computer and every piece of software to create a galactic comparison
graph i also have a 3 month old baby and am about to move house
I do hope to do some similar testing with Sonar in a few weeks
and also try a Core 2 Duo for comparison, which won't probably be as dramatic.
Cheers
Robin
--------------------
http://www.pc-music.com
|
Martin Walker
Watcher Of The Skies
Joined: 28/02/01
Posts: 12487
Loc: Cornwall, UK
|
Re: Vista performance testing
[Re: robinv]
#434201 - 15/03/07 12:48 PM
|
|
|
Quote robinv:
What i mean is that
doing traditional benchmarking on the system might have shown much closer performance
results, but in reality the XP install was not as good at maintaining a stable CPU usage
of 4 cores in Cubase and so glitched much earlier than the Vista install - you'd only know
that if you were listening to it.
Very good point Robin!
Thanks for all the hard work you've done so far,
and for Vin for dropping in to remind us of his Dawbench tests - I'm a firm believer in
these benchmarks, as I discussed back in PC Notes July 2006:
www.soundonsound.com/sos/jul06/articles/pcnotes_0706.htm
Martin
--------------------
YewTreeMagic
|
Jadoube
member
Joined: 13/05/03
Posts: 346
Loc: Calgary, Canada
|
Re: Vista performance testing
[Re: robinv]
#434490 - 15/03/07 10:09 PM
|
|
|
Hey Robin, great test! Drivers appear to be the biggest stumbling block. My DAW will be
XPP a while yet. I have been using Vista on my laptop and I quite like it... I won't be
going back.
I have one question which you sort of hint at on your web page.
Would you care to comment specifically on improvements you saw by turning off Aero? I
understand if you don't want to give away some tweaker trade secrets, but... I have seen
some opinions that suggest with the correct video card, Aero should actually be more
efficient than the old API. It would seem you found this not to be the case?
--------------------
David
|
TAFKAT
member
Joined: 08/01/03
Posts: 130
Loc: Australia
|
Re: Vista performance testing
[Re: robinv]
#434505 - 15/03/07 10:40 PM
|
|
|
Quote robinv:
On some other
forums my results have taken some crticism but i think it may be down to people thinking
in benchmarks rather than glitch free playback. What i mean is that doing traditional
benchmarking on the system might have shown much closer performance results, but in
reality the XP install was not as good at maintaining a stable CPU usage of 4 cores in
Cubase and so glitched much earlier than the Vista install - you'd only know that if you
were listening to it.
Hey
Robin,
I have been reading some of the fallout , and all I can say is that
those who are speaking the loudest should knuckle down and present some quantifiable data
of their own.., all else is hot air :-)
Your test approach for the plugIns and
polyphony tests are very similar to mine, all have the same methodology of stressing the
system until audio playback is interrupted. I have been a very vocal opponent of VST/CPU
meter benchmarks, as they really do not represent anything past watching a bouncing
meter.., IMO, so I know where your coming from there.. :-)
Anyhow,
On one recent thread at N.com, the debate moved to XP64, where the point was made that
an end user had not seen any benefit with Vista64 over XP64. From my testing, apart from
the obvious advantage of larger memory addressing, XP64 did not show any advantage over
XP32 performance wise. So the performance delta indicated by your testing where there was
a huge difference between XP32 and Vista32/64 is definately a point of discussion. So
either XP64 shows a huge increase over XP32 or we have some conflicting data/opinion,
however I don't expect to get any quantifiable data from the individual involved
Is it possible for you to rerun the test sessions on XP64, just for comparison sake ?
I know, a lot will be screaming , " whats the point, everything is moving to Vista
", and I agree, but it would be great to dot the i's.
I will be getting around
to a XP32/64 v Vista32/64 shootout when I find the time, it is something that is
preciously short of supply at the moment.
Thanks again for sharing the info, it
has definately got the Vista Audio discussion rolling, which will inevitable be a good
thing for all involved.
Peace:
V:
|
table for two
active member
Joined: 24/03/02
Posts: 5808
|
Re: Vista performance testing
[Re: TAFKAT]
#434512 - 15/03/07 11:01 PM
|
|
|
Agreed V
Cant wait to get my laptop choice sorted (dually), so that I can run
Sonar tests on xp32/64 & vista 32/64.
I should really also do it on a
decent enough Centrino, say a 2Ghz one.
Don't have any Turions, bit of a
bummer there.
|
waxwobbler
member
Joined: 28/01/02
Posts: 217
|
Re: Vista performance testing
[Re: robinv]
#434631 - 16/03/07 10:07 AM
|
|
|
Thanks Robin that’s really interesting.
Vista does actually look very nice &
great to know that it may well perform better. My free upgrade is in process after my
recent 2Ghz c2d laptop purchase.
Unfortunately all this technology will force
me to drop Logic 5.5 (don’t laugh). Logic just doesn’t take advantage of the new fast
lappy compared to my 3Ghz P4 where as my Reason instrument count doubled.
Seems
well written software & drivers are key too all this & I guess that means VSTI & effects
throughout.
Thanks again.......
--------------------
www.myspace.com/colossaldj www.7161.com/~colossal
|
robinv
Joined: 31/08/04
Posts: 277
|
Re: Vista performance testing
[Re: TAFKAT]
#434698 - 16/03/07 12:16 PM
|
|
|
Quote TAFKAT:
Is it
possible for you to rerun the test sessions on XP64, just for comparison sake ?
I have done tests on
earlier systems with XP x64 and seen some good improvements over WinXP 32 on a Core 2 Duo
system. And my initial testing of Vista 64 on the same system showed XP x64 ahead on
performance. I was using a different system and a different audio interface. The
difficulty i have is that x64 introduces another level of complexity and confusion into
the game when what i really want it simplicity. In the first instance you have a 64bit OS
upon which Cubase is not supposed to show any improvement - but it does. Secondly in light
of Vista 64 it's unlikely that any more hardware development will be done on x64 and so
the small amount of compatible stuff is not going to grow making it as difficult to
recommend as Vista - but at least Vista will improve.
So, for these tests, i
wanted to avoid over complicating the matter or getting sidetracked into the 64bit debate.
However, everyone has their own angle, dreams and desires and i guess i'll never fulfill
them all
When i do the next load of testing (my test system is curently being borrowed)
i'll try to widen the field a bit in terms of systems and OS used.
Amongst some
of the severe responses to these results there have been some really useful discussions,
thoughts and ideas, so thanks for that - it's all good stuff.
--------------------
http://www.pc-music.com
|
TAFKAT
member
Joined: 08/01/03
Posts: 130
Loc: Australia
|
Re: Vista performance testing
[Re: robinv]
#434730 - 16/03/07 01:31 PM
|
|
|
Quote robinv:
So, for
these tests, i wanted to avoid over complicating the matter or getting sidetracked into
the 64bit debate.
Hey
Robin,
I can understand where your coming from, however with the results of
Vista32/64 being identical, it does throw up the question whether the same behaviour is
being witnessed between XP32/64 or whether there is an improvement. Judging by your
earlier experiences of XP64 v Vista64, it would suggest the equivalent performance gain
from XP32 to XP64 that you experienced to Vista64. That is something I did not experience,
so it looks like I'll need to go back and re dot the i's.
Where you using C4
in the earlier testing as well ?
I also understand that XP64 was side stepped
by the majority of audio developers, which is quite disappointing , as the Core of the
XPx64 OS is actually W2K3-64 , which has a more robust and advanced code base. For those
needing to access more than the current 2 GB per App in XP32 - or 3GB with the .ini switch
, XPx64 could be a good stepping stone before diving head first into Vista..
Peace:
V.
|
robinv
Joined: 31/08/04
Posts: 277
|
Re: Vista performance testing
[Re: TAFKAT]
#434906 - 16/03/07 07:12 PM
|
|
|
Quote TAFKAT:
Judging by your
earlier experiences of XP64 v Vista64, it would suggest the equivalent performance gain
from XP32 to XP64 that you experienced to Vista64.
Ummmm not entirely sure if i understand you. In my older testing
(unpublished) XP64 did outperform XP32 on the same machine. So that's different to my
Vista testing which showed with those components, with that software, there was no
difference.
--------------------
http://www.pc-music.com
|
TAFKAT
member
Joined: 08/01/03
Posts: 130
Loc: Australia
|
Re: Vista performance testing
[Re: robinv]
#434936 - 16/03/07 08:45 PM
|
|
|
Quote robinv:
Quote TAFKAT:
Judging by your
earlier experiences of XP64 v Vista64, it would suggest the equivalent performance gain
from XP32 to XP64 that you experienced to Vista64.
Ummmm not entirely sure if i understand you. In my older testing
(unpublished) XP64 did outperform XP32 on the same machine. So that's different to my
Vista testing which showed with those components, with that software, there was no
difference.
Sorry Mate,
I can see
where the confusion hit..
That should read..
"Judging by your
earlier experiences of XP64 v Vista64, it would suggest the equivalent performance gain
from XP32 to XP64 that you experienced from XP32 to Vista64 "
Peace:
V:
|
Phat Riffioso
Joined: 05/08/03
Posts: 498
Loc: London
|
Re: Vista performance testing
[Re: TAFKAT]
#435405 - 18/03/07 12:39 PM
|
|
|
did someone mention benchmarks?
i had look at the L-factor II bench in vista
and XP yesterday.
here are my results
All latest drivers and updates
as of 17/03/07
Fireface 800
E6600 2.ghz 266fsb, 667 spd Ram, P5w-DH board
Xp
64 - 0 additional notes of polyphony
128- 60
256 - 130
512
-170
1024 -180
Vista
64 - 0 additional notes of polyphony
128-
0
256 - 40
512 -120
1024 -150
Vista wihout aero
64 - 0
additional notes of polyphony
128- 0
256 - 80
512 -150
1024 -170
Vista wihout aero and cubase's priority set to very high
64 - 0 additional
notes of polyphony
128- 70
256 - 130
512 -170
1024 -190
Overclocked 3.29ghz, 366fsb, 915mhz SPD
Xp
64 - 100 additional notes of
polyphony
128- 230
256 - 290
Vista
64 - 30 additional notes of
polyphony
128- x (test wouldn't run)
256 - x
Vista wihout aero
64 - 40 additional notes of polyphony
128- x
256 - x
Vista wihout
aero and cubase's priority set to very high
64 - 80 additional notes of polyphony
128- 200
256 - 280
So performance is lower with this benchmakr on a Dual
Core system in Vista. I have a feeling that cubase SX3 is partly to blame since it's
multiprocessor support in Vista seems quite flaky. several of the tests wouldn't run
without multiprocessing taken off as indicated by the X's above. Even though the X test
had drop outs, less drop outs occured with multiprocessing turned off.
Process
priority also had to be turned up to it's maximum to get the best out for cubase and even
then tiny little background processes would cause drop outs in the audio.
Obviously Cubase SX3 isn't supported under Vista and i am certain that software writen
for Vista would make better use of mutliprocessors and process priorities. i'm guessing
cubase 4 would run better.
I also think that windows aero will have less effect
on audio performance once better grpahics drivers are written. My card was a 7300gt.
--------------------
Kasha - Picture a beautiful life
|
TAFKAT
member
Joined: 08/01/03
Posts: 130
Loc: Australia
|
Re: Vista performance testing
[Re: Phat Riffioso]
#435549 - 18/03/07 09:44 PM
|
|
|
Quote PhatRiff:
So
performance is lower with this benchmark on a Dual Core system in Vista. I have a feeling
that cubase SX3 is partly to blame since it's multiprocessor support in Vista seems quite
flaky. several of the tests wouldn't run without multiprocessing taken off as indicated by
the X's above. Even though the X test had drop outs, less drop outs occured with
multiprocessing turned off.
Hey PhatRiff,
Thanks for sharing, Mate.. :-)
From those results, the Vista Performance gains definitely seem to be application
dependent. Did you manage to run Blofelds across the system on Vista as well, that will
give us a good indication what is happening there.
Also, very interesting find
on the Multi Processor /Priority settings. Seems SX3/N3 is not a definate go on Vista..,
its also interesting that it was at the over clocked settings you experienced the reported
issues.
You wouldn't have a copy of XP64 handy, would you ?
I'll
have some fuel for the fire soon..
Peace:
V:
|
Phat Riffioso
Joined: 05/08/03
Posts: 498
Loc: London
|
Re: Vista performance testing
[Re: TAFKAT]
#438223 - 24/03/07 01:40 PM
|
|
|
No xp64 here I'm afraid. Only got Vista32 as well since it's OEM.
I hope to
test Cubase 4 side by side but i'm being put off by the current stability problems. i
don't see many C4 benchmarks about?
I also hope to try blofelds soon but it's
pretty time consuming.
Good report here. btw http://www.aavimt.com.au/dawbench/blofelds-ocotocore.htm
--------------------
Kasha - Picture a beautiful life
|
Rowboffin
new member
Joined: 22/05/03
Posts: 429
|
Re: Vista performance testing
[Re: robinv]
#438242 - 24/03/07 02:48 PM
|
|
|
Great stuff Robin and PhattRiff.
Have either of you tried turning UAC off in
these benchmarks to see if there's a performance penalty involved?
|
robinv
Joined: 31/08/04
Posts: 277
|
Re: Vista performance testing
[Re: Rowboffin]
#438621 - 25/03/07 06:13 PM
|
|
|
Quote Rowboffin:
Great stuff
Robin and PhattRiff.
Have either of you tried turning UAC off in these
benchmarks to see if there's a performance penalty involved?
UAC was off in the tweaked Vista tests - not
sure why there'd be a performance hit, it's just a user security facility (or is it
something else?). It's probably the most annoying feature ever invented if, like me, you
are installing/uninstalling and moving things around all the time - however, if it's a
shared PC or you just want to use the software and prevent yourself from accidentally
messing things up then it's fab!
Why do you think they'd be a performance
penalty?
--------------------
http://www.pc-music.com
|
Rowboffin
new member
Joined: 22/05/03
Posts: 429
|
Re: Vista performance testing
[Re: robinv]
#438897 - 26/03/07 12:23 PM
|
|
|
I've read this report of some severe application slowdowns in a disk intensive application
here.
They could easily be specific to the application been discussed, but since its a
disk-intensive application it triggered my curiosity about whether a similar effect may
occur in a DAW.
UAC is one of the things about Vista I have some trepidation
about, although to an extent I understand the motivation behind it. For security on
Windows to improve we certainly need to move towards a situation where, for the majority
of the time, we run as standard users and not administrators.
|
Martin Walker
Watcher Of The Skies
Joined: 28/02/01
Posts: 12487
Loc: Cornwall, UK
|
Re: Vista performance testing
[Re: Rowboffin]
#439374 - 27/03/07 11:16 AM
|
|
|
I know UAC can be annoying, but like Robin I doubt that it would impose a measurable
performance penalty.
Personally I think it's well worth accepting the
inconvenience rather than disabling the UAC - after all, as I reported back in PC Notes
December 2004, I once lost all my Waves plug-in authorisations when my system clock
mysteriously jumped forward by a year, and UAC would have prevented this happening!
Martin
--------------------
YewTreeMagic
|
TAFKAT
member
Joined: 08/01/03
Posts: 130
Loc: Australia
|
Re: Vista performance testing
[Re: robinv]
#452259 - 26/04/07 07:59 AM
|
|
|
Hey All,
I finally managed to knuckle down and do the Shootout between XP32 v
XP64 v Vista 64, using Blofelds DSP40 and L-Factor II benchmarks in N3 ..
Following Robins comments regarding the performance of XP64 being on par to Vista 64, I
wanted to see if those performance findings correlated to the specific benchmarks I had
developed..
I have posted these in a 2 part series at DAWbench,and Nuendo, I'll
summarise for the members here...:-)
Here we go..
---------------------------------------------------------
Blofelds DSP 40R5 :
Core2Quad QX6700 @ 2.66 GHZ : 1066 FSB : Nuendo 3.2
Motherboard : i975
Memory : 2 GB - XMS 6400 @ PC6400 - 800 MHZ : Standard Timings by SPD :
O.S :
XP64 SP1 :
Lynx 2 : Driver : V2 Build 14
256 Samples - 166 Magneto's
/ Save-ReOpen 166 :
128 Samples - 142 Magneto's / Save-ReOpen 142 :
064 Samples - 107 Magneto's / Save-ReOpen 107 :
032 Samples - 64 Magneto's /
Save-ReOpen 55 :
---------------------------------------------------
Compared to the XP32 results
Blofelds DSP 40R5 : Core2Quad QX6700 @
2.66 GHZ : 1066 FSB : Nuendo 3.2
Motherboard : i975
Memory : 2 GB -
XMS 6400 @ PC6400 - 800 MHZ : Standard Timings by SPD :
O.S : XP SP2 :
Lynx 2 : Driver : V2 Build 13n
256 Samples - 165 Magneto's / Save-ReOpen 165
:
128 Samples - 142 Magneto's / Save-ReOpen 142 :
064 Samples - 108
Magneto's / Save-ReOpen 108 :
032 Samples - 70 Magneto's / Save-ReOpen 60 :
--------------------------------------------------------
Very much on
par down to 064 , slightly down on the 032 results.
Moving onto L-Factor II
:
----------------------------------------------------------
L-Factor II:R9 : Core2Quad QX6700 @ 2.66 GHZ : 1066 FSB : Nuendo 3.2
Motherboard : i975
Memory : 2 GB - XMS 6400 @ PC6400 - 800 MHZ : Standard
Timings by SPD :
O.S : XP64 SP1 :
Lynx 2 : Driver : V2 Build 14
064 Samples - 380 Added Notes of Polyphony:
032 Samples - 250 Added
Notes of Polyphony:
----------------------------------------------------------
Compared to
XP32
L-Factor II:R9 : Core2Quad QX6700 @ 2.66 GHZ : 1066 FSB : Nuendo 3.2
Motherboard : i975
Memory : 1 GB - XMS 6400 @ PC6400 - 800 MHZ :
Standard Timings by SPD :
O.S : XP SP2 :
Lynx 2 : Driver : V2 Build
13n
064 Samples - 450 Added Notes of Polyphony:
032 Samples - 310
Added Notes of Polyphony:
-------------------------------------------------------------
Well that is a
bit of a surprise in light of some reports of increased performance being achieved under
XP64 compared to XP32 with N3.
I could have actually nudged them up a bit
higher, but I was getting an occasional part dropouts, it wasn't as smooth and linear as
it was under XP32... :-(
So for those contemplating a trip to XP64land
With neither N3/C4 having 64 bit register /memory addressing capabilities , as much as I
like the overall feel of XP64 , I can't suggest any viable reason to be making the move
from XP32.
On to Vista64..
------------------------------------------------------------
Blofelds DSP
40R5 : Core2Quad QX6700 @ 2.66 GHZ : 1066 FSB : Nuendo 3.2
Motherboard :
i975
Memory : 2 GB - XMS 6400 @ PC6400 - 800 MHZ : Standard Timings by SPD :
O.S : Vista64 Business :
Lynx 2 : Driver : V2 Build 14
256
Samples - 160 Magneto's / Save-ReOpen 160 :
128 Samples - 130 Magneto's /
Save-ReOpen 130 :
064 Samples - 087 Magneto's / Save-ReOpen 079 :
032 Samples - No Go From Get Go :
-------------------------------------------------
Moving onto L-Factor
II :
-------------------------------------------------
L-Factor
II:R9 : Core2Quad QX6700 @ 2.66 GHZ : 1066 FSB : Nuendo 3.2
Motherboard :
i975
Memory : 2 GB - XMS 6400 @ PC6400 - 800 MHZ : Standard Timings by SPD :
O.S : Vista64 Business :
Lynx 2 : Driver : V2 Build 14
064
Samples - 290 Added Notes of Polyphony:
032 Samples - No Go From Get Go :
---------------------------------------------------
Well that's an
anti-climax if ever I saw one.. LOL
The behaviour of Nuendo running the test
was also extremely erratic, constant pauses, some up to 20-25 seconds long, I kid you not,
sluggish, more farping and parping than either XP32 or XP64.., the real time audio engine
aspect of Nuendo under Vista took an absolute pounding..:-(
BTW: I used the
Aero interface that everyone so dearly loves, simply because its supposed to lighten the
CPU loadings due to it offloading graphics directly to the GPU.., simply scrolling caused
audio drop outs !!
The install was tweaked to as high a level as I could, while
still remaining Aero active.. , in short, the only thought I had was.. "What a piece of
shite "... LOL !!
After reading Rains initial report, I was optimistic, after
doing my own testing I am more Misty Optic.. :-(
Perhaps C4 , which was used by
Rain in their testing , has some more native optimisation in regards to Vista, the 2
current tests are not really suitable for C4, so that will remain a grey area for now.
The other grey area, is what do Rains Vista Optimisations actually entail past my
own tweaks.., I honestly can't see what they could possibly have done to return such a
huge performance improvement.
These tests are in no way conclusive in regards
to DAW performance under Vista , I just wanted to dot the i's using the current
applications and benchmarks.
As far as I am concerned , I wouldn't be touching
it with a barge pole until all applications/plugins are ported natively across.. , in
other words, talk to me in 2008.. :-)
Peace:
V:
|
Martin Walker
Watcher Of The Skies
Joined: 28/02/01
Posts: 12487
Loc: Cornwall, UK
|
Re: Vista performance testing
[Re: TAFKAT]
#452358 - 26/04/07 11:17 AM
|
|
|
Thanks for all the hard work Vin.
Re XP64 - I know some people are reporting
significant hikes in performance, but others are suggesting that it's the 64-bit code of
the application being used that's better optimised than the same application in its 32-bit
version, rather than any inherent advantage in 64-bit per se.
Martin
--------------------
YewTreeMagic
|
robinv
Joined: 31/08/04
Posts: 277
|
Re: Vista performance testing
[Re: TAFKAT]
#452543 - 26/04/07 04:23 PM
|
|
|
That's a good load of new data mate - nice one.
Here's some of my thoughts in light
of why my experience might have been different.
First off, you're using a Quad
core processor. One of the things that seemed to come to light in my testing was that XP
really doesn't like four cores, so maybe that's contributing to the lack of performance
difference between XP and XP64. Personally on a dual core system i have seen a good
advantage to XP64 - but then i test with Cubase 4.
Which brings me to Nuendo 3
- Steinberg say it's not currently compatible with Vista http://knowledgebase.steinberg.de/173_1.html
so you could
probably expect it not to give spectacular results, or at least be a bit weird - it might
explain why it didn't do very well.
Aero - this is a major hog and reduces
performance no end. Not sure who told you it "lightens the CPU load" but they are out of
their tree. We ran all our tests with "Vista Basic" graphics which worked nicely
thanks.
It's really hard trying to compare computers using different software
and different configurations because the permutations are huge. I would recommend trying
Cubase 4 though, or Sonar for that matter - something that is Vista compatible and maybe
make up your own tests as i appreciate the benchmark stuff you used isn't in Cubase 4
format and is a pain in the arse to adjust.
Cheers
Robin
--------------------
http://www.pc-music.com
|
TAFKAT
member
Joined: 08/01/03
Posts: 130
Loc: Australia
|
Re: Vista performance testing
[Re: robinv]
#452705 - 26/04/07 09:41 PM
|
|
|
Quote robinv:
Here's some
of my thoughts in light of why my experience might have been different.
First
off, you're using a Quad core processor. One of the things that seemed to come to light in
my testing was that XP really doesn't like four cores, so maybe that's contributing to the
lack of performance difference between XP and XP64. Personally on a dual core system i
have seen a good advantage to XP64 - but then i test with Cubase 4.
Hey Robin,
Interesting
findings on Quadcores, I have had no performance issues with XP32/64 with the Quadcores at
all, in fact , they scaled as expected. So you are saying if using a Dualcore, I would
have achieved better scaling between XP32/XP64 ?
Quote:
Which brings me to
Nuendo 3 - Steinberg say it's not currently compatible with Vista http://knowledgebase.steinberg.de/173_1.html
so you could
probably expect it not to give spectacular results, or at least be a bit weird - it might
explain why it didn't do very well.
I did make a note that the performance increases you reported could be due to
Cubase4 having added optimisations for Vista , that is the grey area I cannot comment on
as yet, and until Steinberg come clean with what they are, if they in fact exist, I'll
reserve comment :-)
Re Nuendo 3 not being Vista compatible, hmmm, thats
Steinbergs way of saying , if you want to play, you are on your own.., they also state
Cubase 4 isn't Vista 64 compatible.. :-)
N3 isn't officially XP64 compatible
either, nor is Cubase 4 more optimised for XP64 over XP32.., where you claim you saw some
performance advantage , so we still have some i's to dot.. :-)
Quote:
Aero - this is a major
hog and reduces performance no end. Not sure who told you it "lightens the CPU load" but
they are out of their tree. We ran all our tests with "Vista Basic" graphics which worked
nicely thanks.
The
statement in regards to the Aero lightening CPU load was not something I ascribe to, but
in response to various members at Nuendo believing that due to Aero rendering directly to
the GPU. The information is directly from Microsoft and various other reports I have read
on the web.
My Aero experience was bare minimum , in fact it was basically
indentical to Vista basic, no transparency, very minimal eye candy, etc, etc. I will run
the benches again with Vista Basic just to dot the i's further.., I will be very surprised
if they show any significant improvement..
Quote:
It's really hard trying to compare computers using
different software and different configurations because the permutations are huge. I would
recommend trying Cubase 4 though, or Sonar for that matter - something that is Vista
compatible and maybe make up your own tests as i appreciate the benchmark stuff you used
isn't in Cubase 4 format and is a pain in the arse to adjust.
I hear you Mate, and that is why I
specifically stated that these results were not conclusive. I just wanted to round out
some testing with the current benchmarks before I move to the next round. I won't bother
creating anything for C4 atm, as I believe it to be simply a stop gap release. To be
honest I think it is nothing more than a polished turd, no 64 bit register or memory
addressing, SSE4 optimisations, etc, etc..
Anyhow,
Thanks again
for all of your input on all of this :-)
I'll drop back later today with
the Vista Basic results just to round this off..
Peace:
V:
|
robinv
Joined: 31/08/04
Posts: 277
|
Re: Vista performance testing
[Re: TAFKAT]
#452734 - 26/04/07 11:01 PM
|
|
|
XP64/XP32 scaled as expected? So you weren't expecting any difference? Oh - i would have,
but only because that's been my experience.
x64 compatibility - (from another
Steinberg kb) Cubase SX 3, Nuendo 3 and Cubase (Studio) 4 are the first versions of
Steinberg DAW products to be compatible with a 64-bit operating system.
So it
should at least benefit from overall system performance increases running 64bit - this is
why i thought the Quad might be causing the trouble.
Vista compatibility - if
you look at the article Cubase 4 sits in a box marked "Windows Vista compatible products"
whereas Nuendo sits in a box marked "Windows Vista compatible products after future
update". So i think Steinberg is being as clear as they can while they come up with fully
fledged native support.
Personally i think Cubase 4 is fabulous - best update
in years, but then it takes all sorts 
--------------------
http://www.pc-music.com
|
TAFKAT
member
Joined: 08/01/03
Posts: 130
Loc: Australia
|
Re: Vista performance testing
[Re: robinv]
#452737 - 26/04/07 11:14 PM
|
|
|
Hey Robin,
Just to round this off..:-)
Here are the Vista Basic
results..
--------------------------------------------------------
Blofelds DSP 40R5 : Core2Quad QX6700 @ 2.66 GHZ : 1066 FSB : Nuendo 3.2
Motherboard : i975
Memory : 2 GB - XMS 6400 @ PC6400 - 800 MHZ : Standard
Timings by SPD :
O.S : Vista64 Business :
Lynx 2 : Driver : V2 Build
14
256 Samples - 165 Magneto's / Save-ReOpen 165 :
128 Samples - 134
Magneto's / Save-ReOpen 134 :
064 Samples - 104 Magneto's / Save-ReOpen 100
:
032 Samples - No Go From Get Go :
-------------------------------------------------
Moving onto L-Factor
II :
-------------------------------------------------
L-Factor
II:R9 : Core2Quad QX6700 @ 2.66 GHZ : 1066 FSB : Nuendo 3.2
Motherboard :
i975
Memory : 2 GB - XMS 6400 @ PC6400 - 800 MHZ : Standard Timings by SPD :
O.S : Vista64 Business :
Lynx 2 : Driver : V2 Build 14
064
Samples - 350 Added Notes of Polyphony:
032 Samples - 220 Added Notes of
Polyphony:
--------------------------------------------------
Definitely an improvement over the tweaked Aero configuration , especially on L-Factor
II.., still not up to XP64 tho.
What was really interesting to note was the
difference in overall system behaviour, feeling a lot more fluid while running the tests,
no long pauses as when using the stripped back Aero, so there is definately a resource hit
when using Aero, even when it is stripped back. This could be also dependent on Video
cards being employed. I am using smaller workstation based Nvidia Quadro cards on my
systems, which wouldn't be high on the Vista rating..:-)
Either way,
Not a huge improvement as you reported with C4, but enough non the less, I'll be
definitely avoiding Aero from now on..
Thanks Mate.. :-)
Peace:
V:
|
TAFKAT
member
Joined: 08/01/03
Posts: 130
Loc: Australia
|
Re: Vista performance testing
[Re: robinv]
#452769 - 27/04/07 03:51 AM
|
|
|
Quote robinv:
XP64/XP32 scaled as
expected? So you weren't expecting any difference? Oh - i would have, but only because
that's been my experience.
Hey Robin,
Sorry didn't see this post till way after I posted the other..
I said the Quadcores scaled as expected, in response to your claim that the
Quadcores are not scaling correctly under XP32/64. In regards to XP32 v XP64, using SX/N3,
there is absolutely no reason for either app to be able to deliver better performance
under XP64. If you have experienced a quantifiable scalability with SX/Nuendo with XP64
over XP32, please post the results.. ;-)
Quote:
x64 compatibility - (from another Steinberg kb)
Cubase SX 3, Nuendo 3 and Cubase (Studio) 4 are the first versions of Steinberg DAW
products to be compatible with a 64-bit operating system.
So it should at least
benefit from overall system performance increases running 64bit - this is why i thought
the Quad might be causing the trouble.
Steinberg claiming that the applications are compatible with
XP64,is simply stating that they will install and run on the O.S, as will most 32 bit
applications.. :-)
There is no 64 bit register or memory addressing in
N3/SX3/C4, so the code base has absolutely no benefit in running on a 64 bit O.S, how can
there be ?, Something you yourself confirmed with your Vista32/64 bit results.
Quote:
Vista compatibility -
if you look at the article Cubase 4 sits in a box marked "Windows Vista compatible
products" whereas Nuendo sits in a box marked "Windows Vista compatible products after
future update". So i think Steinberg is being as clear as they can while they come up with
fully fledged native support.
Steinberg have a habit of talking in diagonalese at the best of times, you actually use
that as any kind of tangible reference.. ??
LOL, so what makes Cubase 4 more
compatible with Vista ?
BTW: The Nuendo 3 install simply required a double
click , the Cubase 4 install required an optional "Vista Only" installer that needed to be
downloaded to even be able to install the application. The standard installer stating that
Cubase 4 required XPSP1 or higher.. ??
Yeh I can see how much homework
Steinberg put into C4 being Vista ready ?? LOL :-)
Quote:
Personally I think Cubase 4 is fabulous - best
update in years, but then it takes all sorts
We do find our own truth...
:-)
V:
|
robinv
Joined: 31/08/04
Posts: 277
|
Re: Vista performance testing
[Re: TAFKAT]
#452842 - 27/04/07 08:55 AM
|
|
|
Quote TAFKAT:
We do
find our own truth... :-)
Yes we do 
--------------------
http://www.pc-music.com
|
Phat Riffioso
Joined: 05/08/03
Posts: 498
Loc: London
|
Re: Vista performance testing
[Re: robinv]
#453533 - 28/04/07 05:10 PM
|
|
|
The test that i have done with Cubase SX 3 and Vista also bought me to the conclusion that
it wasn't fully compatible with Vista. However, i think that there is still some
development to be done with Audio Interface drivers and Vista Compatibility. For example
simply resizing WMP when playing an Mp3 will cause glitches on my Fireface. Switching to
the fully compatible on-board audio (Realtek HD on asus p5wDH) will run absolutely
smooth.
Now RME claim that the fireface is compatible with vista but i don't
believe this is true. However before i jump to conclusions there is also the possibility
that it is just bad IE1394 drivers for the TI chipset.
--------------------
Kasha - Picture a beautiful life
|
TAFKAT
member
Joined: 08/01/03
Posts: 130
Loc: Australia
|
Re: Vista performance testing
[Re: Phat Riffioso]
#453586 - 28/04/07 10:26 PM
|
|
|
Quote Phat Riffioso:
The test
that i have done with Cubase SX 3 and Vista also bought me to the conclusion that it
wasn't fully compatible with Vista. However, i think that there is still some development
to be done with Audio Interface drivers and Vista Compatibility.
Hey Phatriff,
Definitely still a
few grey areas to be navigated.. :-)
For now, I am sticking to XP32/64/W2K3,
while at the same time, being a glutten for punishment, toying with testing Windows Server
"Longhorn" Beta 3. I have always liked the feel of the Server based O.S's when used as
Workstation platforms.., stripped back, fuid and far more responsive out of the box.. IMHO
:-)
Peace:
V:
|
Phat Riffioso
Joined: 05/08/03
Posts: 498
Loc: London
|
Re: Vista performance testing
[Re: TAFKAT]
#456109 - 03/05/07 10:21 PM
|
|
|
There is some hope. I ran my onboard HD audio with the latest release of ASIO4ALL and
performance wasn't that bad at all. Very close to XP maybe between 5% and 10% slower
across all latency settings. Aero was disabled and indexing turned off. No other tweaks
enabled.
I tried running ASIO4All with the fireface but it can't handle it's
hardware buffers so unfortunately no comparison can be done there. it's a shame because i
would of like to see if the firewire bus was causing problems.
Anyway from all
this what i have been able to work out is that cubase SX3 is a lot more stable under vista
than i initially thought. There still seems to be some interference from background tasks
such as hard disks and services but i'm sure this can be cured with software updates.
Judging by my systems responsiveness and performance under vista it seems unlikely
that i will get any magical boost to processor efficiency in comparison to XP. XP seems to
be the most daw friendly OS there is when it comes to PC's at the moment. Robin posted
some promising scores with a quad core under vista but i'm taking these results with a
pinch of salt until the data can be verified by another party.
What i do find
interesting and hopeful is that ASIO4ALL is wave RT compatible but emulates ASIO. I think
it is quite possible for manufactures to release their own wave RT compatible drivers that
support ASIO. Maybe we will see better performance from these as the driver will be
matched to the hardware unlike the generic approach of ASIO4ALL.
--------------------
Kasha - Picture a beautiful life
|
TAFKAT
member
Joined: 08/01/03
Posts: 130
Loc: Australia
|
Re: Vista performance testing
[Re: robinv]
#456161 - 04/05/07 02:39 AM
|
|
|
Hey PhatRiff,
I have ported Blofelds DSP40 and L-Factor over to Cubase4, and
will be retesting over the next few days. I'll post the results for XP32 v XP64 v Vista /
Nuendo3 v C4 ASAP.
Lets see where the dust settles ..:-)
V:
|
TAFKAT
member
Joined: 08/01/03
Posts: 130
Loc: Australia
|
Re: Vista performance testing
[Re: robinv]
#456608 - 05/05/07 02:24 AM
|
|
|
Hey All,
I have finalised the DAW O.S shootout..
This will be last
series of testing until Nuendo 4.1 / Cubase 4.1
After the last round of tests ,
we still had the grey area of whether C4 was more optimised for Vista, which could explain
the variables in my results over those presented by Rain Recording. I ported the test to
C4 and cranked the identical development system into gear across the 3 O.S's one more
time.
Lets see the results for C4 :
-----------------------------------------------------------------------------
Blofelds DSP 40R5 : Core2Quad QX6700 @ 2.66 GHZ : 1066 FSB : Cubase 4.023
Motherboard : i975
Memory : 2 GB - XMS 6400 @ PC6400 - 800 MHZ : Standard
Timings by SPD :
O.S : XPSP2:
Lynx 2 : Driver : V2 Build 14
256 Samples - 161 Magneto's / Save-ReOpen 161 :
128 Samples - 131 Magneto's /
Save-ReOpen 131 :
064 Samples - 093 Magneto's / Save-ReOpen 093 :
032 Samples - 053 Magneto's / Save-ReOpen 042 :
---------------------------------------------------------------------------
Moving onto L-Factor II :
----------------------------------------------------------------------------------
L-Factor II:R9 : Core2Quad QX6700 @ 2.66 GHZ : 1066 FSB : Cubase 4.02
Motherboard : i975
Memory : 2 GB - XMS 6400 @ PC6400 - 800 MHZ : Standard
Timings by SPD :
O.S : XPSP2 :
Lynx 2 : Driver : V2 Build 14
064 Samples - 370 Added Notes of Polyphony:
032 Samples - 230 Added
Notes of Polyphony:
-----------------------------------------------------------------------------
Blofelds DSP 40R5 : Core2Quad QX6700 @ 2.66 GHZ : 1066 FSB : Cubase 4.023
Motherboard : i975
Memory : 2 GB - XMS 6400 @ PC6400 - 800 MHZ : Standard
Timings by SPD :
O.S : XP64SP1:
Lynx 2 : Driver : V2 Build 14
256 Samples - 162 Magneto's / Save-ReOpen 162 :
128 Samples - 131
Magneto's / Save-ReOpen 131 :
064 Samples - 092 Magneto's / Save-ReOpen 092
:
032 Samples - 051 Magneto's / Save-ReOpen 040 :
---------------------------------------------------------------------------
Moving onto L-Factor II :
----------------------------------------------------------------------------------
L-Factor II:R9 : Core2Quad QX6700 @ 2.66 GHZ : 1066 FSB : Cubase 4.02
Motherboard : i975
Memory : 2 GB - XMS 6400 @ PC6400 - 800 MHZ : Standard
Timings by SPD :
O.S : XP64SP1 :
Lynx 2 : Driver : V2 Build 14
064 Samples - 330 Added Notes of Polyphony:
032 Samples - 140 Added
Notes of Polyphony:
-----------------------------------------------------------------------------
Blofelds DSP 40R5 : Core2Quad QX6700 @ 2.66 GHZ : 1066 FSB : Cubase 4.023
Motherboard : i975
Memory : 2 GB - XMS 6400 @ PC6400 - 800 MHZ : Standard
Timings by SPD :
O.S : VistaBus64:
Lynx 2 : Driver : V2 Build 14
256 Samples - 160 Magneto's / Save-ReOpen 160 :
128 Samples - 125
Magneto's / Save-ReOpen 125 :
064 Samples - 085 Magneto's / Save-ReOpen 080
:
032 Samples - No Go From The Get Go :
---------------------------------------------------------------------------
Moving onto L-Factor II :
----------------------------------------------------------------------------------
L-Factor II:R9 : Core2Quad QX6700 @ 2.66 GHZ : 1066 FSB : Cubase 4.02
Motherboard : i975
Memory : 2 GB - XMS 6400 @ PC6400 - 800 MHZ : Standard
Timings by SPD :
O.S : VistaBus64 :
Lynx 2 : Driver : V2 Build 14
064 Samples - 300 Added Notes of Polyphony:
032 Samples - 060 Added
Notes of Polyphony:
------------------------------------------------------------------------------------
Well there you have it.., if
anything it is at least consistent.
C4 is consistently worse than N3 across
all 3 O.S's, also the performance scaling on Vista is consistent to that of N3, proving
that there is no Vista optimisation in C4, as I had suspected.
This raises a
few serious questions in regards to the results and claims presented by Rain.
1: " Vista outperforms XP without any trouble at all"
This is something that
I could not quantify in any of my extensive testing, I found the exact opposite applies.
XP clearly and easily out performs Vista.
2: Multithreading
capability in XP is not optimised for Quad Cores and could have swayed my earlier results
using N3.
This is something that has absolutely no basis. I have always
experienced a linear and progressive scalability using Single Quadcore and Dual Dualcore
based systems using XP32. I experienced similar scalability in XP64 and Vista64 using both
N3 and C4. If there is any advantage with Vista in regards to better scalability per core,
then it would only come into play above 4 cores. Unfortunately with current state of the
Steinberg MP capability, that answer will remain unconfirmed for now.
3: N3 is
not Vista compatible, whereas C4 is officially supported.
My testing shows that
C4 is no more Vista compatible than N3/SX3, if anything, and to top it off, N3 actually
out performed C4 on Vista on these tests.
So that brings us to the Performance
delta's reported by Rain, and how or why such a delta could have been achieved.
We can rule out O.S, we can also rule out Application , which really only leaves
Hardware.
I used the same reference hardware across 3 different Operating
systems, and did not achieve anywhere near the performance deltas reported , so unless
Rains tests were somehow swayed by the Reality Distortion Field that Apple were once
infamous for, we have a serious conflict in reported results.
The only
explanation I can even allude to is that the RME Fireface totally tanked on XP, and
performed substantially better on Vista. That however is only reflective of the
performance of the audio interface and the respective drivers, not any performance
advantage of the Operating System itself .
In short , my only conclusion is
that the whole Vista Performance report is nothing more than hyperbole designed to
generate some traffic to the respective website , and hopefully hoodwink the unsuspecting
into believing that Rain have somehow discovered the magic formula that no other
Professional DAW builder can match in regards to DAW performance under Vista..
Yes, I admit I am a professional DAW builder, however I use minimal to no self promotion
on forums, and I keep the DAWbench project totally independent of my commercial sites.
The focus has been to present accurate objective information that can be easily
verified and quantified by anyone who is willing to take the time to download and run the
tests for themselves. There are no smoke and mirrors , everything is totally transparent,
and open to scrutiny and debate.
I am more than happy to be proven wrong..
V:
|
jcschild
Joined: 06/07/05
Posts: 295
|
Re: Vista performance testing
[Re: robinv]
#456731 - 05/05/07 03:37 PM
|
|
|
HI Vin,
thanks for all that. your test results parellel mine.
i have
not posted the results as we are trying to do a comprehensive report on what does and does
not work and what actually sees a benefit.
1) with Vista 32 or 64 (whats the
point to 32?) Dual core does NOT show an improvement but rather a decrease.
(opposite of what Robin was saying)
2) Vista did show slightly better scaling with
Quad or Dual Quad (again opposite to Robins comments)
with the Multicore
issue with Cubase/N3 (8 cores) we validated this with Sonar, Reaper as well.
thanks again for all the hard work!
Scott
ADK
Edited by jcschild (05/05/07 03:38 PM)
|
robinv
Joined: 31/08/04
Posts: 277
|
Re: Vista performance testing
[Re: TAFKAT]
#457305 - 07/05/07 08:32 PM
|
|
|
Nice job man - meaty
Of course it raises questions about my results - and so it
should, it's healthy. Hopefully it will stimulate more testing, more discussion about what
actually happens under Vista rather than all this nonsense about what could or what should
happen which is exactly what prompted me to do some testing in the first place.
I think you credit me with far too much sneakiness - it's all much simpler than some big
conspiracy. I was frustrated by the lack of facts in the Vista debate and so, while i was
testing out my first Quad machine i thought i'd invest a day in doing some concentrated
Vista testing. At the same time at Rain we'd put together a "Vista Watch" page where we
were keeping an eye on drivers to see what music hardware/software was compatible - useful
for us, useful for others. Anyway, i did the tests and got amazing results - did it again
and decided it would be useful stuff to go on the Rain website - so that's where it went.
I then came back here to post a link to the results because i said that i would do so.
Did it create any interest and traffic? Good god yes, that link was taken and
posted all over the world (in our tiny niché market world at least) because no one else
had done any testing of music software on Vista of any sort anywhere. What surprised me
even more was how little anyone actually read of it and how quickly people jump to their
own conclusions. Since then i've had to put up with endless accusations and nonsense about
being in Bill Gates pocket, or that i'm just trying to con people into something or other
- i don't get it, i really don't - people are just funny i guess.
As for
challenging my conclusions - sure, but then it's simply what i concluded. In my test Vista
did outperform XP on that machine, with that hardware, under those conditions. From
my tests it was reasonible to conclude that maybe Vista handled the Quad core better than
XP - it made sense and considering that no one else was offering any other information it
seemed reasonible to me - and still does. As i stressed in the results - they are
particular to that system and certainly now that you've introduced new tests on a
different machine with a different person recording results we are likely to see different
things. Maybe we have different tollerances for error - i recorded the XP result at the
first sign of a glitch and i found the results disappointing compare to Dual Core. Who
knows - there are so many factors.
The most important thing is that now many
more people are producing test results and people who are wondering whether to go to Vista
have some good information to go on. As a system builder it is so much better to be able
to advise a customer from personal experience rather than spouting a load of assumed
rhetoric gathered off the internet - if nothing else that makes this whole thing
worthwhile from my point of view.
Now we've got some contradictions i want to
do a load more testing - should i post the results here or is that further market
manipulation?
Scott - not sure what you're on about mate, you've got everything backwards. Both
of your comments are exactly what i've been saying
For me at least i can say with confidence that with the right
hardware and software Vista can work for audio - which is kind of what we were trying to
find out really and your results back that up. Whether it's better or worse than XP is
obviously a different matter and one that would bear further scrutiny. This is complete
déja vú for me of the Win98/XP crossover but hey, there's nothing new under the
sun....Lots more goodies to come i hope.
--------------------
http://www.pc-music.com
|
Previous
Topics
Next
Threaded
Edit
Reply
Quote
