Quantcast
My Windows 7 vs. Windows 8 DAWBench Results - Page 2 - Gearspace.com
The No.1 Website for Pro Audio
My Windows 7 vs. Windows 8 DAWBench Results
Old 13th September 2012
  #31
Gear Addict
 
🎧 5 years
Thank you for this! love this board
Old 13th September 2012
  #32
Lives for gear
 
1 Review written
🎧 10 years
Quote:
Originally Posted by cschormann ➡️
The Cakewalk team posted a comparison on their blog that indicates just the opposite - strong performance gains across the board with Windows 8 running Sonar X1, as compared to Windows 7.

It is possible that Cubase and Sonar use the OS and the hardware differently, but I guess that that is not the issue here. I think I have to try this for myself - I am a Cubase user

See this blog post from Cakewalk:
The Cakewalk Blog

And a summary of their findings:

SONAR CPU load at low latency

SONAR CPU gains were observed when using Windows 8 for Low latency performance tests. These gains mean you can run bigger loads in Win8 at low latency without audio glitching.

low latency plugins… 15.5% CPU reduction
input monitoring… 8% CPU reduction
high track count… 23% CPU reduction
High bandwidth audio …6.2% CPU reduction

SONAR multi-core CPU load balancing at low latency

Workloads for cores are more evenly balanced at low latencies on Windows 8. Better balanced core workloads translate to more efficient use of multiple CPU core hardware and thereby better workload scaling for large projects.

low latency plugins… 23% improvement
input monitoring… 31.7% improvement
high track count… 30.6% improvement
High bandwidth audio …17.5% improvement

Memory usage

A 7.9% reduction in memory use under Win8 was observed when loading a large real world SONAR project (Cori Yarkin project from SONAR sample content) under identical system configuration. Reduced memory load can be observed in most of the tests.

Disk Performance

A 78% improvement under Win8 was observed in disk read/write performance while reading large buffer sizes. Improvements were more moderate at smaller buffer sizes.

System calls per second

An 85% reduction in system calls was observed under Windows 8 in the input monitoring case and more moderate gains in the other cases. Fewer system calls translate to improved CPU load as well as fewer user mode to kernel transitions which mean fewer audio glitches.

Kernel use

25 – 50% reduction in kernel use can be observed in some of the tests with Win8. Lower kernel use results in fewer audio glitches since it leaves more headroom for audio drivers.
Now I know I have to test cubase...

Sent from my PC36100
Old 13th September 2012 | Show parent
  #33
Lives for gear
 
TAFKAT's Avatar
 
🎧 15 years
Quote:
Originally Posted by cschormann ➡️
The Cakewalk team posted a comparison on their blog that indicates just the opposite - strong performance gains across the board with Windows 8 running Sonar X1, as compared to Windows 7.
Those results don't prove anything IMO, as I noted in an earlier post , they simply represent a performance meter metric that they use internally in development. The % variance is in the meter only , which may or may not have any correlation to the actual scaling performance in Real World application.

It all sounds very technical and convincing to some, but I read it and picked up very early on that unless the metric can be applied across multiple DAW's, it is of little value past hype.

Those same developers also claimed strong performance gains in Vista over XP , how did that pan out ?

I have also seen the Cake team post various claims of huge % improvements of multiprocessor x-scaling / low latency performance, which amounted to a hill of beans when placed comparatively against the other DAW's.

Lets see where the dust settles when SONAR is placed side by side against Cubendo/Reaper/StudioOne.

Old 14th September 2012
  #34
Gear Addict
 
🎧 10 years
The motivation behind the rigged performance test from Cakewalk was they have a partnership with Microsoft to promote new Windows OS. They did it with Vista- Windows 8. To say it is biased is an understatement. Not saying 8 is good or bad either way but you can safely ignore their flawed test method as any indication of Windows 8 performance.
Old 14th September 2012 | Show parent
  #35
Gear Nut
 
🎧 10 years
"The motivation behind the rigged performance test from Cakewalk was they have a partnership with Microsoft to promote new Windows OS. They did it with Vista- Windows 8. To say it is biased is an understatement. Not saying 8 is good or bad either way but you can safely ignore their flawed test method as any indication of Windows 8 performance. "

Windows 7 is the best operating system for pro audio and it shown in all the benchmark tests performed on this site. Why would they need Cakewalk to lie on benchmarks? Windows 8 with my Motu MK3 is identical to windows 7 using Studio one. I thinks its worth the $40 upgrade and the more pro audio guys upgrade the more people will bug MS to release a Midi WinRT driver(Which the said they were working on).
Old 15th September 2012
  #36
Gear Addict
 
🎧 10 years
You say identical, they say huge performance increase for audio apps. Just saying it was more a Windows 8/Cakewalk commercial than an unbiased test. Running at 1024 buffers won't tell a thing because most issues are exposed when running lower latency. Anyway, we will see a more clear picture once more people put it through it's paces with Dawbench.
Old 17th September 2012 | Show parent
  #37
Lives for gear
 
Psychlist1972's Avatar
 
🎧 5 years
This thread got me to finally create an account here. I've been a lurker for a long time.

Without getting to any of the other stuff raised here, I'm curious if wireless was on during the tests?

If so, any chance of running the Win8 test again with wireless disabled? You can get a lot of latency if there are conflicts with that, and sometimes they don't show up until an OS reinstall.
DPC Latency Checker

This can be especially true if you have any apps actually using wireless during that time.

On my Windows 7 machine, I had that issue early on with the regular network adapter on my Gigabyte UD9 board. The impact is based in part on how the interrupts are hard-wired in the machine's board. Wireless tends to be an issue more often, especially on notebooks.

Pete
NOTE: I work for Microsoft, but participate here mainly in a personal capacity. I will, however, pass any relevant data on this topic back to the product teams.
Old 18th September 2012
  #38
Lives for gear
 
TAFKAT's Avatar
 
🎧 15 years
Quick heads up.

I am in the final stages of configuring updated sessions for SONAR X1 to bring it in line with the other DAW applications that have the full series of tests available. I had abandoned SONAR as a reference DAW after some curves encountered with the development team a few years back and the lack of interest from the SONAR community , but this has motivated me to get it back up to date.

Sessions that will be available :

DAWbench DSP - RXC/RXC-EXT , MD5, EMP, CSP, TTCC, TLA

DAWbench VI - CV/NCV

Details Here

All the DSP sessions are completed , just finishing off the VI sessions and I will then upload them so that they are freely available to everyone so that they can do their own testing.

We can then get a better idea how the Cakewalk W8 DAW Meter results correlate to the DAWbench results.

Old 18th September 2012 | Show parent
  #39
Lives for gear
 
Sounddesigner's Avatar
 
🎧 10 years
Quote:
Originally Posted by TAFKAT ➡️
Quick heads up.

I am in the final stages of configuring updated sessions for SONAR X1 to bring it in line with the other DAW applications that have the full series of tests available. I had abandoned SONAR as a reference DAW after some curves encountered with the development team a few years back and the lack of interest from the SONAR community , but this has motivated me to get it back up to date.

Sessions that will be available :

DAWbench DSP - RXC/RXC-EXT , MD5, EMP, CSP, TTCC, TLA

DAWbench VI - CV/NCV

Details Here

All the DSP sessions are completed , just finishing off the VI sessions and I will then upload them so that they are freely available to everyone so that they can do their own testing.

We can then get a better idea how the Cakewalk W8 DAW Meter results correlate to the DAWbench results.


Vinn, have you considered benchmarking Sonar X2 wich will be released at the end of this month? Sonar X2 is suppose to have improved Low-Latency performance and other audio-engine improvements. X2 just seems like a better choice then X1 for benchmarking (but it would also be nice to know how much of a low-latency improvement took place from X1 to X2). I seen your old benchmarks for Sonar 7 when Cakewalk claimed performance improvements and your benchmarks showed that improvement to be extremely small. Some Sonar users like myself are interested in your benchmarks and feel they are needed to keep companies more honest and or open the eyes of users.


EDITED
Old 18th September 2012 | Show parent
  #40
Lives for gear
 
TAFKAT's Avatar
 
🎧 15 years
Just another quick heads up.

Porting the DAWbench VI session to SONAR is doing my head in, I have some embedded control that is zeroing the volumes in Kontakt on stop, so it is essentially ignoring all volume pan information in the mixer. I can grab a fadar and move it a touch and it will jump to the correct volume, but each time I hit stop it zero's the volume again in Kontakt , ignoring the fadar position.

Sorted, don't ask me how tho.

I deleted all of the Kontakt instances and redid all of the track assignments to the appropriate Kontakt instances/ parts and all the volume information sorted itself out.

I still have one issue tho , with the other DAW's polyphony is assigned dynamically, so as soon as I unmute a sustained part the polyphony is added in Kontakt, with SONAR it is only added on the initial attack which is at the start of the track only, so polyphony is only applied on loopback.

Might be a setting I'm missing.. ?


Last edited by TAFKAT; 18th September 2012 at 03:56 AM.. Reason: Sorted first issue..still needing some guidance.
Old 18th September 2012 | Show parent
  #41
Lives for gear
 
TAFKAT's Avatar
 
🎧 15 years
Quote:
Originally Posted by Sounddesigner ➡️
Vinn, have you considered benchmarking Sonar X2 wich will be released at the end of this month? Sonar X2 is suppose to have improved Low-Latency performance and other audio-engine improvements. X2 just seems like a better choice then X1 for benchmarking (but it would also be nice to know how much of a low-latency improvement took place from X1 to X2). I seen your old benchmarks for Sonar 7 when Cakewalk claimed performance improvements and your benchmarks showed that improvement to be extremely small. Some Sonar users like myself are interested in your benchmarks and feel they are needed to keep companies more honest and or open the eyes of users.
Hey SD,

Absolutely, I will be benching SX2 as soon as I can get my hands on it.

All of the work configuring the sessions in SX1 will still be of value to do a direct head to head between SX1 and SX2 as well as the other competing DAW's.
Old 18th September 2012 | Show parent
  #42
Lives for gear
 
Sounddesigner's Avatar
 
🎧 10 years
Quote:
Originally Posted by TAFKAT ➡️
Hey SD,

Absolutely, I will be benching SX2 as soon as I can get my hands on it.

All of the work configuring the sessions in SX1 will still be of value to do a direct head to head between SX1 and SX2 as well as the other competing DAW's.

OK, That's good to hear! Looking forward to seeing the results. I bought the Pre-sale X2 and am waiting for it and will test it with my music projects but your benchmarks i'm definitely interested in. Cakewalk refuses to answer questions regarding the low-latency improvements on their forum (even in a thread they were envolved in answering other audio-engine questions but not the low-latency performance questions wich the thread was all about - strange). I bought X2 largely because of the claimed low-latency improvement (There is no demo for X2 and by the time a demo is released the price will significantly increase for me so i had to gamble by purchasing it now and hope Cakewalk is telling the truth about the performance increase. But now seeing Cakewalk ignoring questions regarding the low-latency performance does'nt look good!).

I'm a lover of audio-engine improvements and do hope Cakewalk is telling the truth about X2 since i don't like being lied to and suckered in. I know they have stretched the truth in the past so maybe i'm being foolish for giving them some trust. I do wish strongly for Sonar to be improved in the Audio-Engine department (better low-latency, gapless audio engine, core scaling improvements) as well as other users. I don't know why DAW makers are slow to bring such improvements (Well i suspect it takes enormous coding to accomplish, but still strong desire should be there to do it. But also it probably does'nt bring in the money like more shiny new plugins and sample library's do.). Hope the Windows 8 claims are true as well. Anyways, thanks for doing this, and again i'm looking forward to the results!


EDITED
Old 18th September 2012 | Show parent
  #43
Lives for gear
 
TAFKAT's Avatar
 
🎧 15 years


This is a screen grab of SONAR X1 running the DAWbench VI NCV template at 032 samples. I have highlighted the disproportional spiked core/thread that is still inherent with the audio engine , so I am not placing too much weight in their low latency claims.. :-(

I haven't participated over at their forms for years and don't plan to again to be honest, so I am not really keeping abreast of the chatter. I am not overly surprised that they are avoiding the questions they deem in the too hard basket ... :-)

Old 18th September 2012 | Show parent
  #44
Lives for gear
 
TAFKAT's Avatar
 
🎧 15 years
Just tidying up some final sessions and I'll get to work and do some comparative shoot outs across multiple DAW's - Cubase / Reaper / SONAR / StudioOne.

Each have their quirks but I have to tell ya StudioOne thru a curve at me with the DAWbench VI sessions as I was putting some finishing touches to them. I had the session ported by a trusted colleague more intimate with SO2 and hadn't really had a closer look until yesterday. I went to the Mixer Console and scratched my head trying to find the dedicated faders for the respective MIDI channels, the session after all is focussed on large numbers of MIDI tracks assigned to multiple instances of Multi Timbral Virtual Instruments, i.e 6 x K4 running 16 parts in each , so needing to access the volume / pan of those tracks individually is .., hmmm, important.

Well , heres the thing , StudioOne does not support faders/pan/console entries for MIDI tracks, nope, nada, none , so obviously anyone using Multi Timbral instruments whether they be VI's or External are going to have a juggle on their hands.

I have to be missing something, surely... ?

I digress...

New test sessions will be uploaded and available to all later this week

Stay Tuned.

Old 19th September 2012 | Show parent
  #45
Gear Addict
 
🎧 10 years
Don't use Sonar but wondering did you try limiting Sonar from using 1-2 cores with core affinity? I know some DAW's deal with multi processing different and you have to shut down cores for it to work right.
Old 19th September 2012 | Show parent
  #46
Lives for gear
 
TAFKAT's Avatar
 
🎧 15 years
Quote:
Originally Posted by Gear Geek ➡️
Don't use Sonar but wondering did you try limiting Sonar from using 1-2 cores with core affinity? I know some DAW's deal with multi processing different and you have to shut down cores for it to work right.
I have never seen any discussion from the developers suggesting that the area of core affinity should be investigated. The spiked thread being displayed does not correlate to the TM/Cores , well not to the same degree, its an internal indicator.

Having said that, when that spiked thread hits red, the engine stops , no matter if you have substantial resources still available.
Old 19th September 2012
  #47
Here for the gear
 
🎧 5 years
I personally have had great results with my recent upgrade Windows 8 is heaven for me .... On Windows 7 I could not even use more then 3 rtas Vi instances at once on Windows 8 i can run 30 instances at a low buffer here is my post I posted at the DUC forum


"So I just got a copy of Windows 8 for being enrolled as an IT student & I have heard that there are always minor bugs with new OS's but I took the gamble and did a clean install of Windows 8 & Pro Tools & FL Studio ..... ALL I got to say is it runs so smooth Windows 8 seems a little bit lighter then 7.

Last night I was doing a stress test seeing how far I could go with out CPU errors: I had 20 different instances of RTAS VI's going at once no tweaks needed & it was only going up to like 35% max on the system usage meter and thats with me recording to a 5400rpm usb 3.0 wd passport & not my 7200rpm drive I was at 128 samples 3 of 4 processors and 85% cpu

SO if you got Pro Tools 8 LE an are scared to upgrade when Windows 8 drops youll be fine go for it I love it so far it seems really stable and has me excited to get back to work on my latest projects"

https://pbs.twimg.com/media/A296m4wCAAAujHX.jpg:large < Snapshot

Ill have to do a bench mark test for my laptop

as of now I love my new setup like 0 latency and no more limitations
Old 21st September 2012 | Show parent
  #48
I've been away from the board for a while due to my work schedule, so let me catch up here...

Quote:
Originally Posted by norbury brook ➡️
Did you do a clean install of windows 8 or an upgrade ?

I've tested windows 8 on my x58 DAW and it performed very well using cubase 6 x64 and all x 64 plugins,VI's etc so I was surprised at you results. I did a clean install though on a fresh drive.
My first round of tests was with an upgrade installation. Started with a production Win 7 image (minus some apps I knew wouldn't survive the upgrade), and then upgraded to Win 8 RTM. In just a few moments, however, I will post results based on my production Win 7 install and a totally fresh Win 8 install.

And note that Windows 8 performs fine with my personal projects as well. I only started to see difference in projects that contained literally hundreds of plugins and stacked polyphonic notes.

Quote:
Originally Posted by Psychlist1972 ➡️
Without getting to any of the other stuff raised here, I'm curious if wireless was on during the tests?

If so, any chance of running the Win8 test again with wireless disabled?[/url]
Ask and you shall receive! (Actually I did this last weekend, because I wanted to rule that out too.) Anyway, I did indeed run some tests without my PCI wireless adapter installed- and also with a much more modest graphics card than my normal one. This delivered a slight performance boost in the PCIe tests I did, but not to the USB or FireWire tests (as expected). However, the performance change applied to both operating systems, and didn't close the gap between the OS results. So I do not believe WiFi has anything to do with my results.

Quote:
Originally Posted by Shizzl3Mayn3 ➡️
I personally have had great results with my recent upgrade Windows 8 is heaven for me .... On Windows 7 I could not even use more then 3 rtas Vi instances at once on Windows 8 i can run 30 instances at a low buffer here is my post I posted at the DUC forum
Your Win 7 setup had serious problems. Even a really chumpy PC can normally handle much more. Whatever issue you encountered was apparently resolved once you had Win 8 set up.

But again, using Win 8- even in a DAW- doesn't feel any different from Win 7. My results come from tests that push my audio hardware to its absolute limits. Most people never push their systems this hard in day-to-day use.
Old 21st September 2012 | Show parent
  #49
The Windows 7 vs. Windows 8 DAW benchmarks I did a week ago left me with more questions than answers. There was an obvious performance gap between the two operating systems in my tests, but none of my data gave me any hints as to why. In order to gain some more clarity on the issue (and satisfy my GAS), I picked up a few new interfaces of different kinds and ran some new tests.

The interfaces I tested for this round are:
  • MOTU PCIe-424 with 24I/O expander. (PCI Express)
  • MOTU UltraLite-mk3 pre-hybrid (FireWire 400)
  • Native Instruments Komplete Audio 6 (USB)
  • Virus TI Desktop interface mode (USB)

** PREPARATION **
In last week’s tests, I started out with my production DAW Windows 7 image on my original 500GB hard drive. After doing my Win 7 tests, I upgraded that image to Windows 8 for the second round.

I made several changes to my PC since then that necessitated re-testing my UltraLite results, so here’s what’s different between the tests I ran then and now:
  • I did not uninstall Microsoft Security Essentials or Acronis True Image before doing the Windows 7 tests, so Windows 7 had a little more of a resource burden this time around. (Turns out those programs result in around a 6% performance penalty for Windows 7. More on that later.)
  • This time I was running Windows 8 as a fresh install on a separate partition with only the minimum drivers and apps required for the tests.
  • These tests were all done with a new 2TB hard drive as the system drive.
  • Because of some equipment shuffling I did last week, I had to switch from using a 6-foot FireWire cable to a 15 foot one for the UltraLite tests. The 24I/O also came with a 15-foot cable.
  • Due to some new hardware I purchased there were more drivers and services loaded on both operating systems this time.

So, I expected (and observed) both some slightly reduced performance results versus last week’s on the UltraLite, and a slightly smaller gap between the Win 7 and Win 8 rounds.

** THE TESTS **
I performed each test at least once for each interface, on each operating system. If I encountered results that were better or worse than expected, I rebooted the computer and re-tested to confirm the results were accurate. Doing this never significantly changed the results. Just like last week the tests were the following.
  • CSP - Instances of URS Classic Console Strip Pro
  • MD5 - Instances of Wave Arts MultiDynamice 5
  • NCV - Notes of polyphony without effects on Kontakt 4
  • CV - Notes of polyphony with convolution reverb on Kontakt 4

RAW DATA



PER-INTERFACE RESULTS

24I/O


KOMPLETE AUDIO 6


ULTRALITE


VIRUS


** OBSERVATIONS **
  • The FireWire and PCIe interfaces saw a similar performance drop on Win 8 as on my previous tests. As I expected, the gap wasn’t as big because my Win 7 partition had more apps and services loaded than the Win 8 partition. The difference was still significant- a 21%-44% drop-off in performance on Windows 8.
  • The two USB interfaces fared significantly better on Windows 8, in a sense. The Komplete Audio 6 only suffered a 4-12% drop, and the Virus lost only 0-4%- and even scored 2% higher on the CSP test!
  • The goal of every test is to find the highest number of plugin instances or notes of polyphony without any audible pops or dropouts in the audio tracks of the project. With the PCIe and FireWire interfaces, if you were at that maximum, enabling one additional plugin or one more row of polyphony introduced maybe one pop every 3-4 seconds or so. Enabling a couple more would result in pops maybe every 2-3 seconds perhaps, and so on- but it took a while before you ended up with a simply unplayable result. The “no pops” threshold with these interfaces generally happened well before Cubase’s ASIO meter “redlined.” But with the USB interfaces, I could drive the ASIO meter all the way up into the red without hearing any pops- however, adding just a single plugin or polyphony track beyond that threshold would result in completely unplayable results; harsh buzzing, silent gaps, etc.
  • While USB seemed to have a higher playable threshold than FireWire and USB in terms of ability to perform under severe ASIO strain on Windows 8, the environment was much more fragile. For example, while playing at the “no pops” threshold on the PCIe/FireWire, you could tab around, launch programs, etc without interrupting the audio, but with the USB interfaces, even something as trivial as launching Notepad.exe- sometimes even just moving the mouse- resulted in severe disruption of the audio.
  • I used a program called RTL Utility, which measures actual round-trip latency on audio interfaces, giving you more accurate numbers than what you would see by looking in your DAW’s audio interface panel. I found that the latency numbers (both reported and real) between the two operating systems were virtually identical. The difference between Win 7 and Win 8 seems to be more a matter of load/bandwidth than raw speed.
  • In an effort to see if PCI/PCIe saturation had anything to do with the poor showing for the MOTU interfaces on Windows 8, I replaced my high-end NVIDIA GeForce GT 570 card with a much more modest GeForce GT 430, and also removed my PCI WiFi card. This improved the PCIe interface tests by around 3.5 percent overall, but had no effect on the FireWire results. Also, the PCIe performance boost occurred on both operating systems, so it didn’t help explain the gap.
  • Due to the differences in my Windows 7 setup between last week’s tests and this week’s tests, my Windows 7 tests in this round were around 6% worse. In other words, the extra software I normally have running for antivirus and backups incur a 6% performance hit. My Windows 8 test results between the two weeks for that interface were basically the same. So keep in mind that my Windows 7 results this week sort of have a 6% handicap. The Virus may have scored the same on both operating systems this week, but if I had uninstalled Microsoft Security Essentials and Acronis True Image, it very likely would be 6% worse on Windows 8. (Which is still much better than the other interfaces.)

** CONCLUSIONS **
It’s still way too early to declare that there’s a serious issue in Windows 8 (and if there is, what exactly is the root cause), but I think we can at least rule some potential culprits out.
  • It’s not Cubase. The TI Virus just about the same on both operating systems (with the caveat that the Windows 7 results have around a 6% handicap). If Cubase were the problem, I’d have expected the performance gap to be consistent across interfaces.
  • It’s not JUST FireWire. Both my PCIe and FireWire interfaces underperformed similarly on Windows 8. Efforts to reduce PCI saturation only slightly improved results for the PCIe interface, and didn’t affect the FireWire interface at all.
  • It’s not necessarily drivers. Both my PCIe and FireWire interfaces are MOTU products, but the two devices have different features, different hardware, and different drivers. It COULD be a case of unoptimized drivers in both cases... but if that’s the case, what’s different about Windows 8 that it would cause these logo-compliant Windows 7 drivers to underperform?
Old 21st September 2012 | Show parent
  #50
Quote:
Originally Posted by [email protected] ➡️
The newest OS is never designed for the last generation computer, and it is always going to eat more resources, just as newer versions of your DAW will have more features and power, but will want a more powerful computer.
I can see where you might develop that line of thinking, but Windows 7 was way more lean than Vista, and people keep saying that Windows 8 is leaner and quicker than Windows 7, for some reason.

Quote:
Originally Posted by TAFKAT ➡️
Just tidying up some final sessions and I'll get to work and do some comparative shoot outs across multiple DAW's - Cubase / Reaper / SONAR / StudioOne.
I am stoked to see this. I almost considered getting SONAR simply for benchmarking, but hearing what a pain it's been for you, I'm glad I passed.
Old 21st September 2012 | Show parent
  #51
Lives for gear
 
TAFKAT's Avatar
 
🎧 15 years
Hey U,

Thanks again for the time and energy.

Some very interesting results that throw up numerous questions in regards to the variances attributed to the protocol/drivers of the respective interfaces.

FWIW: The threshold that you experienced the breakdown with the FW and USB interfaces maybe more to do with the respective interfaces than the protocol itself , but its worth investigating further. There has been changes to the FW stack of Win8 ( I know , here we go again )

I'll add FW/USB to the test pool as well as the reference PCIe card, I'll be using all RME interfaces across the range of latencies HDSPe AIO / Fireface 800 / Babyface , to stay as consistent as possible and to give a wider cross reference to the interfaces you have already tested.

I'll upload all the new sessions over the weekend , I am currently laid up with a heavy flu ( again) , so time and energy is not on my side but I will get the sessions uploaded so that anyone interested can play/follow along.

I'll have to outline the respective curves being thrown up with the DAW Hosts, i.e, SONAR/StudioOne will not dynamically allocate sustained polyphonic voices when unmuting tracks in DAWbench VI , only at reloop.

This will make testing a little more involved as we will need to load up the polyphony and then either stop and restart at each point or wait and allow the arrangement to play thru and loop back at each added polyphonic instance, which will increase the testing time but is necessary to get a comparative.

I'll also use a different range of PlugIns to get a wider test pool of results, most likely RXC-EXT , Softube Tube Tech Classic Channel and Waves Kramer Tape. Those 3 hit the system a lot harder and will be a better option moving forward to the newer X79/Hexacore dev system as the previous range of plugins are getting a little light on for the faster rigs.

I'm very keen to get the SONAR results up to see how they correlate to Cakewalks , which will give us a better indication on how that internal meter of theirs holds up... :-)

Peace
Old 21st September 2012
  #52
I forgot to mention in my previous benchmark post that as before, all tests were performed at 256 samples, with 44.1kHz sample rate, at 32-bits. I recorded both the reported and measured latency of each interface on both operating systems. The reported numbers (as shown in the Cubase interface panel) were identical on both OSes, and the measured numbers, as computer by Oblique Audio's RTL Utility, were very nearly identical.

For reasons I haven't been able to figure out yet, I wasn't able to complete the RTL (round-trip latency) tests on the Virus TI Desktop. That's why there's no data for it on the "actual" test results.

RAW DATA (lower numbers are better)


CHART (shorter bars are better)


That's it for now. Excuse me while I go off to ponder whether a 5.041ms latency advantage is worth the $1200 USD price difference between my PCIe-424 core system and the Komplete Audio 6.
Old 21st September 2012 | Show parent
  #53
Lives for gear
 
TAFKAT's Avatar
 
🎧 15 years
Thanks for the added clarification.

The extremely high Output Latency of the Virus TI helps explain why there was minimal difference comparatively to the other interfaces. There are huge safety buffers being employed there , essentially 5 x above nominal.

I have also noticed the RTL Utility can have issues when such extended safety buffers are employed.

Lets see where the dust settles once we push the latencies down.

Old 22nd September 2012 | Show parent
  #54
Lives for gear
 
Sounddesigner's Avatar
 
🎧 10 years
Quote:
Originally Posted by TAFKAT ➡️
I have never seen any discussion from the developers suggesting that the area of core affinity should be investigated. The spiked thread being displayed does not correlate to the TM/Cores , well not to the same degree, its an internal indicator.

Having said that, when that spiked thread hits red, the engine stops , no matter if you have substantial resources still available.

In Sonar you can Edit the Audio.ini file for better multi-core performance with computers that have more cores and threads. The default model is 1, you'll want to choose 2. Go to Edit>Preferences>Advanced (at the bottom of preferences box)>Audio>Configuration file>There on the right there is all the settings, scroll down to> ThreadSchedulingModel- the range is from
<0 - 2>, Choose 2 then save it (verify change took place).

For more info go to help files in in Sonar X1 and do a search for Audio.ini then Scroll down to ThreadSchedulingModel, It states : This variable goes in the [Wave] section and controls the interaction of the main audio thread and worker threads on multiprocessor systems when the Use Multiprocessing Engine option is enabled. Depending on the system, a particular model may result in less glitching and better overall performance. The values are as follows:
0 = Same as previous versions of SONAR.
1 = (default) Better thread balance. Model is more efficient and can provide cycles for other tasks.
2 = Additional worker thread is created. This may result in improvement with Quad processor systems or higher. Not recommended for Dual processor systems.

I know it's possible you already knew this info, just thought i'd post it just incase otherwise.
Old 22nd September 2012 | Show parent
  #55
Lives for gear
 
TAFKAT's Avatar
 
🎧 15 years
Hey SD,

I had already covered the area of editing the Audio.ini with the devs back in the day, and my question then and still is, why is it necessary ?

I would expect that by default the audio engine should already be optimised for higher core/thread counts.

The spiked thread is the "worker thread" from my understanding.

I'll test SONAR with the default settings as the devs did in their reports.

I am not really interested in circumnavigating SONAR's inherent idiosyncrasies to be honest , the whole area of multiprocessor optimisation is a discussion I exhausted with the devs a long time ago.

I Digress,

Getting back to the Win7/Win8 shootouts.

I have encountered issues with both my Waves and iLok licensing on Win8 , so the testing is going to be even more tedious than I had planned.

Waves new USB licensing is not working at all, iLok2 is scanning licenses extremely slowly , so imagine loading a session with 320 instances... :-(

Of course raising a support query will result in the usual and valid dismissal - Windows 8 is not officially released or supported... :-(

I'll stay on it.

Quick Edit : Sorted the iLok2 issue by reinstalling the latest driver.. Phew... :-)


Last edited by TAFKAT; 22nd September 2012 at 01:28 AM.. Reason: Sorted iLok2 issue
Old 22nd September 2012 | Show parent
  #56
Lives for gear
 
Sounddesigner's Avatar
 
🎧 10 years
Quote:
Originally Posted by TAFKAT ➡️
Hey SD,

I had already covered the area of editing the Audio.ini with the devs back in the day, and my question then and still is, why is it necessary ?

I would expect that by default the audio engine should already be optimised for higher core/thread counts.

The spiked thread is the "worker thread" from my understanding.

I'll test SONAR with the default settings as the devs did in their reports.

I am not really interested in circumnavigating SONAR's inherent idiosyncrasies to be honest , the whole area of multiprocessor optimisation is a discussion I exhausted with the devs a long time ago.
OK, i understand your position on this. I agree also that the 'secret tweaks' approach should'nt be necessary.
Old 22nd September 2012 | Show parent
  #57
Lives for gear
 
Psychlist1972's Avatar
 
🎧 5 years
Quote:
Originally Posted by TAFKAT ➡️
Hey SD,

I had already covered the area of editing the Audio.ini with the devs back in the day, and my question then and still is, why is it necessary ?
Not defending them, but given fast enough processor, the optimum number of threads in any application is 1. Maybe the config setting *forces* the additional threads, which would be bad in a single or even dual core machine.

It even took Adobe until pretty recent versions to properly support multi threading and multi processor setups in their video encoding.

That said, I haven't seen a single core computer in quite a while, and the only dual core I have is a tablet.

Pete
Old 22nd September 2012 | Show parent
  #58
Lives for gear
 
TAFKAT's Avatar
 
🎧 15 years
Quote:
Originally Posted by Psychlist1972 ➡️
It even took Adobe until pretty recent versions to properly support multi threading and multi processor setups in their video encoding.
Video Editing developers have been notoriously slow at embracing SMT/SMP compared to DAW developers. The DAW developers are about 5 years ahead in some cases.

This is not exactly new ground for DAW's.

Sonar X2 downloading as I type , it will be interesting to see what they have delivered ( if anything ) over X1 in regards to LLP and X-Scaling.
Old 22nd September 2012 | Show parent
  #59
Motown legend
 
Bob Olhsson's Avatar
 
🎧 15 years
Quote:
Originally Posted by Psychlist1972 ➡️
Not defending them, but given fast enough processor, the optimum number of threads in any application is 1. ...
That matches what friends who are developers told me when I was deciding between a slower 8 core and a faster 4 core computer. I was told the point of diminishing returns for a single application is 4 which would make 5 ideal assuming all other factors are equal. The others are the number of DSP calculations per cycle which multiplies the effective clock rate and the speed of the RAM buffer.
Old 22nd September 2012 | Show parent
  #60
Lives for gear
 
TAFKAT's Avatar
 
🎧 15 years
Quote:
Originally Posted by Bob Olhsson ➡️
That matches what friends who are developers told me when I was deciding between a slower 8 core and a faster 4 core computer. I was told the point of diminishing returns for a single application is 4 which would make 5 ideal assuming all other factors are equal. The others are the number of DSP calculations per cycle which multiplies the effective clock rate and the speed of the RAM buffer.
Hey Bob,

How long ago was that , if you don't mind me asking ?

True, X-Scaling has never been linear in regards to return per core/threads, i.e 1 to 2 , 2 to 4 , 4 to 8 , 8 to 16 , so laws of diminishing returns are still valid in regards to DAW performance, but less significant in the current architectures than what was experienced even a few gen back. Clock speed will always remain a significant factor , but the relevance of weighing up faster clock over numbers of cores in bang for buck decisions is not as relevant with the current architectures as it was , say 4-5 years ago when clock speed / cores were more of a premium than today.

None of the above has any baring in SONAR's comparative inefficiencies in SMP.

The current X79/Socket 2011 platform is a good test bed for comparing the x-scaling within a single architecture from 4 to 6 to 8 cores on a single socket system , which I can use to complete the exercise of getting some solid numbers of clock/core in respect to DAW scaling. I have the Hex/Octo numbers with the chips at almost identical base/turbo clockspeeds , the Quad has a few more cycles , but I'll get some numbers when I get a breath and crunch the numbers re % improvements for those interested.

Anyhow,

Swinging this back , I'll have some numbers for Win7/8 on the X58/Nehalem Quad system initially and then eventually I'll run some numbers across the X79 / Sanbybridge-E Hex when I get around to configuring Win8 on that platform as well.. :-)

📝 Reply

Similar Threads

Thread / Thread Starter Replies / Views Last Post
replies: 4616 views: 615184
Avatar for smoke
smoke 7th May 2021
replies: 98 views: 39103
Avatar for dfghdhr
dfghdhr 5th June 2021
replies: 50 views: 45765
Avatar for Eat
Eat 15th March 2013
replies: 1156 views: 196506
Avatar for Al Yeska
Al Yeska 2nd December 2019
Post Reply

Welcome to the Gearspace Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…

Forum Jump
Forum Jump