Jump to content


Photo
- - - - -

ReBench - benchmarking Renoise and your CPU.


  • Please log in to reply
47 replies to this topic

#26 ffx

ffx

    Guruh Motha Fakka is Levitating and Knows Everything About Renoise Member

  • Normal Members
  • PipPipPipPipPipPipPipPipPipPipPipPipPip
  • 2839 posts
  • Gender:Not Telling
  • Interests:Renoise Sidechain

Posted 01 January 2016 - 12:14

aero disabling? really?? Is this still 1995 ?


Edited by ffx, 01 January 2016 - 12:15.


#27 toblerpone

toblerpone

    Guruh Member

  • Normal Members
  • PipPipPipPipPipPipPipPipPipPipPipPip
  • 902 posts
  • Gender:Not Telling
  • Location::c

Posted 01 January 2016 - 12:16

ffx: it ads a small %
or do you think that's cheating? :P

Edited by toblerpone, 01 January 2016 - 12:16.


#28 ffx

ffx

    Guruh Motha Fakka is Levitating and Knows Everything About Renoise Member

  • Normal Members
  • PipPipPipPipPipPipPipPipPipPipPipPipPip
  • 2839 posts
  • Gender:Not Telling
  • Interests:Renoise Sidechain

Posted 01 January 2016 - 12:31

its dos then again. And yes, renoise was made for win not for dos. Its cheating.


Edited by ffx, 01 January 2016 - 12:32.


#29 Abissus

Abissus

    Big GrandDaddy Member

  • Normal Members
  • PipPipPipPipPipPipPipPip
  • 457 posts
  • Gender:Male
  • Location:Terror Australis

Posted 01 January 2016 - 12:58

just push start and type "power options"
 
also turn off your antivirus and everything that might take up cpu processing, you might be able squeeze a bit more by enabling a high contrast theme (disables any aero features as you can't otherwise in 8.1 (maybe you can in 10, I don't know yet) it's also not very clear if it was a retirement, but disabling fpattern follow and opening a non-pattern tab with any meters or scopes not vissible should increase your result by around 20%)

Aah I understand. Something i did when I first got the PC, so no problems there.

I will disable anti virus and the options you suggested.

I can disable Aero via software called "Ultimate Windows Tweaker" (I have begun using it after my last bench), though it seems as this may be "cheating?"

At the moment I have only used it for a few minor things that I doubt will really affect performance, though I am happy to list if necessary. Also If the purpose is to lessen cpu processing, I can use UWT to increase performance further too.


Edited by Abissus, 01 January 2016 - 12:59.

"Music in the soul can be heard by the universe."


#30 ffx

ffx

    Guruh Motha Fakka is Levitating and Knows Everything About Renoise Member

  • Normal Members
  • PipPipPipPipPipPipPipPipPipPipPipPipPip
  • 2839 posts
  • Gender:Not Telling
  • Interests:Renoise Sidechain

Posted 02 January 2016 - 10:41

no, we were joking about "cheating"... I only mentioned because then, windows looks like it was 1995. And I doubt that aero will have such huge impact on performance, although it's pure crapware.

 

I remember another leet windows nerdz haxx:  Set priority to background services (somewhere)... I also remember when I had an and k6, it helped :P

 

You can also install hackintosh of course, if you want to have a REAL computer...


Edited by ffx, 02 January 2016 - 10:41.


#31 toblerpone

toblerpone

    Guruh Member

  • Normal Members
  • PipPipPipPipPipPipPipPipPipPipPipPip
  • 902 posts
  • Gender:Not Telling
  • Location::c

Posted 02 January 2016 - 11:03

"Set priority to background services (somewhere)."

System:Performance:Advanced

But it will only make things worse.



#32 Abissus

Abissus

    Big GrandDaddy Member

  • Normal Members
  • PipPipPipPipPipPipPipPip
  • 457 posts
  • Gender:Male
  • Location:Terror Australis

Posted 03 January 2016 - 03:21


You can also install hackintosh of course, if you want to have a REAL computer...

haha, not likely :P teehee.

 

 

 

no, we were joking about "cheating"... I only mentioned because then, windows looks like it was 1995. And I doubt that aero will have such huge impact on performance, although it's pure crapware.

Aah ok, completely missed that one. :smashed:

Yeah, I haven't really bothered tweaking too much as I haven't felt the need so far.

 

 

 

Have a few things I want to do today, after that I will have another you beaut red hot go at it...

25_biz_kevin.jpg


Edited by Abissus, 03 January 2016 - 03:33.

"Music in the soul can be heard by the universe."


#33 Ledger

Ledger

    Guruh Motha Fakka Knows More About Renoise Than Taktik

  • Normal Members
  • PipPipPipPipPipPipPipPipPipPipPipPipPipPip
  • 3357 posts
  • Gender:Male

Posted 03 January 2016 - 21:26

Thanks for doing this thread!

 

am looking to upgrade, probably to Haswell i7 soon;  I understand it keeps my options more open than Skylake for Windows 7 installs.

 

 

Here are my current results for stock Q6600, Win 7 (no AV disabling, just how I run renoise normally):

 

----------------------------------------------------------------------------------------------

 

Intel Q6600 core2 Quad  2.4Ghz (stock) - 1CPU,latency 10ms   : 11 RPTS
Intel Q6600 core2 Quad  2.4Ghz (stock) - 4CPU,latency 10ms   : 24 RPTS
 
Intel Q6600 core2 Quad  2.4Ghz (stock) - 1CPU,latency 100ms   : 21 RPTS
Intel Q6600 core2 Quad  2.4Ghz (stock) - 4CPU,latency 100ms   : 51 RPTS
 
 
 
So, my 4 cores @100ms looks the same ballpark as a Haswell i7 on 1 core at the same, nice too see! 

Edited by Ledger, 03 January 2016 - 21:30.

--> Lua For Beginners <--
--> Lua for newbies <--

My Scripts On Forum

Top Tip!

 

cpu : Xeon 1231 v3, os : Win 7 64bit, audio: Audient iD4
posts as 4tune @ KvR and some other music related sites


#34 ffx

ffx

    Guruh Motha Fakka is Levitating and Knows Everything About Renoise Member

  • Normal Members
  • PipPipPipPipPipPipPipPipPipPipPipPipPip
  • 2839 posts
  • Gender:Not Telling
  • Interests:Renoise Sidechain

Posted 03 January 2016 - 23:17

Hm a core2quad isnt THAT bad... Did you check the power profile?



#35 toblerpone

toblerpone

    Guruh Member

  • Normal Members
  • PipPipPipPipPipPipPipPipPipPipPipPip
  • 902 posts
  • Gender:Not Telling
  • Location::c

Posted 03 January 2016 - 23:34

Ledger: those results don't seem right at all. Worse case scenario you should be getting around 40-50 for 4CPU



#36 Ledger

Ledger

    Guruh Motha Fakka Knows More About Renoise Than Taktik

  • Normal Members
  • PipPipPipPipPipPipPipPipPipPipPipPipPipPip
  • 3357 posts
  • Gender:Male

Posted 04 January 2016 - 00:48

Not too sure really,

 

power profile set to high performance.  When I upgraded from E6300 core duo I was getting about 3X performance which seemed to agree with Passmark results so I`ve never really questioned it.  Maybe cheap Foxconn motherboard?


Edited by Ledger, 04 January 2016 - 00:48.

--> Lua For Beginners <--
--> Lua for newbies <--

My Scripts On Forum

Top Tip!

 

cpu : Xeon 1231 v3, os : Win 7 64bit, audio: Audient iD4
posts as 4tune @ KvR and some other music related sites


#37 Ledger

Ledger

    Guruh Motha Fakka Knows More About Renoise Than Taktik

  • Normal Members
  • PipPipPipPipPipPipPipPipPipPipPipPipPipPip
  • 3357 posts
  • Gender:Male

Posted 18 January 2016 - 07:01

New Computer Test Yay!

 

Haswell  Xeon 1231 v3   3.4GHz-- similar to i7 without onboard graphics 

 

Interestingly got significant improvements on multi CPU, when I slotted my RAM properly in non-consecutive slots for dual channel operation (16GB; 2X8 Kingston HyperX 1600MHz):

 

 

1) RAM in Single Channel operation

 

1 core 10ms = 45    
8 cores 10ms = 116

 

1 core  100ms = 57

8 cores 100ms = 159 

 

 

2) Dual Channel operation

 

1 core 10ms =  45

8 cores 10ms = 121  (+6 RPTS) 

 

1 core  100ms =  58  (+1 RPTS)

8 cores 100ms = 174  (+15 RPTS)
 

 

3) RAM Dual Channel and Windows 7 64bit set to high performance power mode

 

1 core 10ms = 45

8 cores 10ms = 120

 

1 core 100ms = 58

8 cores 100ms = 199

 

 

 

Minimising renoise GUI seemed to give higher results but not really real world so didn`t bother with full test. 

 

Thanks toblerpone and ffx for the power mode tip!  Got me an extra 25 RPTS @ 100ms!
 


Edited by Ledger, 18 January 2016 - 16:43.

--> Lua For Beginners <--
--> Lua for newbies <--

My Scripts On Forum

Top Tip!

 

cpu : Xeon 1231 v3, os : Win 7 64bit, audio: Audient iD4
posts as 4tune @ KvR and some other music related sites


#38 toblerpone

toblerpone

    Guruh Member

  • Normal Members
  • PipPipPipPipPipPipPipPipPipPipPipPip
  • 902 posts
  • Gender:Not Telling
  • Location::c

Posted 18 January 2016 - 15:10

Could you disable your antivirus and open the phrase tab with the spectrum analyzer disabled (in high performance)?

Thanks.

because I'm thinking of upgrading as well



#39 Ledger

Ledger

    Guruh Motha Fakka Knows More About Renoise Than Taktik

  • Normal Members
  • PipPipPipPipPipPipPipPipPipPipPipPipPipPip
  • 3357 posts
  • Gender:Male

Posted 18 January 2016 - 17:13

Ok here it is:

 

Xeon 1231 v3

 

- Mobo: Asus B85M-G PLUS/USB 3.1 (micro ATX) 

- RAM: 16GB 2x8GB in Dual Channel config Kingston 1600MHz

 

- Win 7 64bit

- Direct Sound

 

- AV = off

- Power mode = high performance

- Phrase Editor focused with no phrase playing

- Renoise master Vu meter is active (horizontal one at the top of the screen)

- No spectrums or osciliscopes

 

 

The (+X) is improvement from my previous best results above

 

90% renoise CPU cut out RPT values:  

-------------------------------------------------
1 core 10ms =  48  (+3)
8 cores 10ms = 127  (+7)

1 core  100ms = 59  (+1)
8 cores 100ms = 202  (+3)

-------------------------------------------------

 

So some improvements but not as major it seems as the dual-channel/high performance mode bumps (though these have not been tested independantly)

 

Also to note it is a fairly fresh Windows install and only critical updates have been installed.  Whether or not non critical updates would improve/ hinder I don`t know.

 

 

Think I`m edged out by the i7`s, maybe we could do with a graph for all the individual best results in this thread! :)   
 


Edited by Ledger, 18 January 2016 - 17:20.

  • toblerpone likes this

--> Lua For Beginners <--
--> Lua for newbies <--

My Scripts On Forum

Top Tip!

 

cpu : Xeon 1231 v3, os : Win 7 64bit, audio: Audient iD4
posts as 4tune @ KvR and some other music related sites


#40 Ledger

Ledger

    Guruh Motha Fakka Knows More About Renoise Than Taktik

  • Normal Members
  • PipPipPipPipPipPipPipPipPipPipPipPipPipPip
  • 3357 posts
  • Gender:Male

Posted 18 January 2016 - 17:47

Also worth noting for Haswell refresh upgrades, some mobos (at least Asus) don`t support the Haswell Refresh chips until a certain BIOS update.  Fortunatly the Asus B85M-G PLUS/USB 3.1 seems to be a late enough board revision, that it worked out of the box.  With the current/ last gen Asus boards you seem to be able to update the bios without posting the CPU via a specified USB port.  However it`s another hoop to be aware of that I`d rather avoid.

 

If unsure and you want to go Haswell the non-refresh chips should be safer with this issue though slightly less performance.

 

If going Skylake you do get more common choices of M.2 onboard slotted ssds.  More possible RAM capacity and DDR4. More choice for USB 3.1 boards  too.   

 

The reason I went with Haswell was no potential problems trying to load up Win 7 via USB ports and a bit lower cost overall which the graphics-less Xeon helped with too.   


Edited by Ledger, 18 January 2016 - 17:54.

  • toblerpone likes this

--> Lua For Beginners <--
--> Lua for newbies <--

My Scripts On Forum

Top Tip!

 

cpu : Xeon 1231 v3, os : Win 7 64bit, audio: Audient iD4
posts as 4tune @ KvR and some other music related sites


#41 Ledger

Ledger

    Guruh Motha Fakka Knows More About Renoise Than Taktik

  • Normal Members
  • PipPipPipPipPipPipPipPipPipPipPipPipPipPip
  • 3357 posts
  • Gender:Male

Posted 18 January 2016 - 18:11

Ok Further Developments, by unpluging my USB 2 Audio interface (Spl Crimson) I was able to get another 14 RPTS on 100ms.  So even on direct sound it looks like audio interface/ card is going to make some difference.  This might explain my poor Q6600 results aswell:

 

So my max score with a completely silent renoise where I can`t see the pattern editor aswell! 

 

is :

 

8 cores 100ms = 216  


Edited by Ledger, 18 January 2016 - 18:15.

  • toblerpone likes this

--> Lua For Beginners <--
--> Lua for newbies <--

My Scripts On Forum

Top Tip!

 

cpu : Xeon 1231 v3, os : Win 7 64bit, audio: Audient iD4
posts as 4tune @ KvR and some other music related sites


#42 ffx

ffx

    Guruh Motha Fakka is Levitating and Knows Everything About Renoise Member

  • Normal Members
  • PipPipPipPipPipPipPipPipPipPipPipPipPip
  • 2839 posts
  • Gender:Not Telling
  • Interests:Renoise Sidechain

Posted 19 January 2016 - 11:56

Well, IMHO firewire is quite a "good" invest currently, since prices for used firewire audio interfaces are massively dropping, thanks to Apple, not supporting it anymore (only officially, yes there is still a firewire driver in OSX 10.11) and trying to force to people to use useless thunderbolt + 20 adapters.

 

A PCIe 1x firewire card with TI chipset is about 10 €. So for around 150-200, you will get an audio interface which did cost over 1000 € some years ago.  Of course have a look first, if there is still a driver for recent Win/OSX.


Edited by ffx, 19 January 2016 - 11:57.


#43 Ledger

Ledger

    Guruh Motha Fakka Knows More About Renoise Than Taktik

  • Normal Members
  • PipPipPipPipPipPipPipPipPipPipPipPipPipPip
  • 3357 posts
  • Gender:Male

Posted 19 January 2016 - 20:46

Well, IMHO firewire is quite a "good" invest currently, since prices for used firewire audio interfaces are massively dropping, thanks to Apple, not supporting it anymore (only officially, yes there is still a firewire driver in OSX 10.11) and trying to force to people to use useless thunderbolt + 20 adapters.

 

A PCIe 1x firewire card with TI chipset is about 10 €. So for around 150-200, you will get an audio interface which did cost over 1000 € some years ago.  Of course have a look first, if there is still a driver for recent Win/OSX.

 

Interesting to know.  It would be good to know what the benchmarks were for different interfaces / connectors too, though were getting into a large number of variables with converters/ drivers etc.  I`ve also got too much invested in the Crimson which has got good AD/DA converters and an analog volume control, which saves speakers/ears from any nasty surprises!.  It`s certainly not perfect as the monitor outputs run quite hot signal wise, so requires further gainstaging after output via monitor/ amp volumes.  Also the volume pot has needed some attention with some DeOxit DN5 spray after it started to crackle badly!, though it seems to be behaving better now.

 

If I was looking again I`d probably be looking at the Audient ID14/22, but unfortunately no analog volume control with those... gladly not looking at the moment for more decisions! head is still spinning trying to get up to date with all the latest pewter stuff!

 

Anyway it seems with these new i7 level chips any losses seem to be diminished.  All my CPU meter anxiety has been really put to bed for a while!  If you really need more power there is always Haswell-E, or if you`ve got money to burn, a high end Xeon E5 or two!


--> Lua For Beginners <--
--> Lua for newbies <--

My Scripts On Forum

Top Tip!

 

cpu : Xeon 1231 v3, os : Win 7 64bit, audio: Audient iD4
posts as 4tune @ KvR and some other music related sites


#44 OopsIFly

OopsIFly

    Guruh Member

  • Normal Members
  • PipPipPipPipPipPipPipPipPipPipPipPip
  • 814 posts
  • Gender:Male
  • Interests:...daydreams... -VS- ...propaganda...

Posted 19 January 2016 - 21:53

Hi, some Linux tests here. Haven't seen some here yet.

 

- Renoise 3.1

 

- Second Hand Workstation, it is one old horse

 

- Xeon w3690 six cores, hyperthreading and turbo boost disabled, so lower single core and not the maximum multicore performance. steampowered.

 

- old m-audio pci sound card

 

- two kernels benched, official ubuntu 4.2.0-lowlatency ("LL") and self-built 4.1.15-rt realtime kernel ("RT")

 

- base is xubuntu 14.04 system, some tweaks applied to ensure better lowlatency operation, but not tuned like a cnc driver

 

- all operation on 64bits

 

- only jackd operation has been measured. alsa prooved weaker and unstable in results - sometimes on par or close to, but randomly giving much lower results if tests are repeated

 

- x-runs occured during tests (sometimes shortly after at the beginning, at pattern 12 or so!), but this seems a renoise-specific problem - I've tested listening to the drone the test produces, and could only hear the real glitches just the moment before hitting the 90% mark

 

- 256 pattern version was used, took ages to load and much much longer to close renoise (but renoise won't crash), it made renoise consume 5.5 gigs of memory.

 

- settings were chosen to the value qjackcontrol presents and that is very close to some digital feedback roundtrip latency I once measured by making a loop in my soundcard that has a hardware mixer in it. "100ms" are buffer settings "2buffers*2048samples@44.1kHz" and "10ms" is "2buffers*256samples@44.1khz". But maybe it is half of the latency just for playback, I'm unsure about it atm, and what comparable settings win/mac people would use.

 

Results:

 

One Core:

"10ms": 39/39 rpts LL/RT

"100ms": 52/53 rpts LL/RT

 

Four Cores:

"10ms": 136/137 rpts LL/RT

"100ms": 183/185 rpts LL/RT

 

Six (full, no HT) Cores:

"10ms": 199/200 rpts LL/RT

"100ms": 86.9%/86.6% @ pattern 257 "tah, linux ruley"

 

 

So you see an old sixcore machine can make teh ruley if all cores are used. And has blow the 256 pattern test. Renoise will properly use any number of cores you feed it with. You will need an even bigger test (or use heavier dsp per channel) for newer generation sixcore cpus, maybe even soon at four-core operation. Also an rt kernel won't be detrimental to the possible workload, but rather improve performance a bit by its better, much more aggressive task scheduling compared to the "soft" lowlatency kernel. Though the reason I installed it was simply to have an option to record drop-out-free at very low latencies (musician monitors mic/instrument with software effects while playing). Maybe I'll try setting the maschine to pure single core operation with turbo boost enabled (OS will see a single core only, too), or enable hyperthreading (12 cores, bwahaha, make amiga ruley even more!) for some fun.


  • Ledger likes this

#45 Ledger

Ledger

    Guruh Motha Fakka Knows More About Renoise Than Taktik

  • Normal Members
  • PipPipPipPipPipPipPipPipPipPipPipPipPipPip
  • 3357 posts
  • Gender:Male

Posted 19 January 2016 - 22:43

Nice!  I didn`t understand half the parameters set up there but that is an impressive end result!

 

 

Shows that  Passmark can be taken with a huge grain of salt with regard to audio performance, though I`m sure your careful optimizations made a big difference too.

 

 

http://www.cpubenchm...1275&cmp[]=2246

 

Intel Xeon W3690 @ 3.47GHz
9,681
Intel Xeon E3-1231 v3 @ 3.40GHz
9,619

--> Lua For Beginners <--
--> Lua for newbies <--

My Scripts On Forum

Top Tip!

 

cpu : Xeon 1231 v3, os : Win 7 64bit, audio: Audient iD4
posts as 4tune @ KvR and some other music related sites


#46 OopsIFly

OopsIFly

    Guruh Member

  • Normal Members
  • PipPipPipPipPipPipPipPipPipPipPipPip
  • 814 posts
  • Gender:Male
  • Interests:...daydreams... -VS- ...propaganda...

Posted 19 January 2016 - 23:44

hm, I've read somewhere a hyperthreading core has about half (or one third) of the throughput of a "real" core. But it is depending a lot on what the other cores do at that moment. That's why I disabled ht on mine - it might speed up rendering, but is no good bet for latency sensitive stuff like realtime audio, and 6 cores are more than enough for me.

 

So your 4/8 cores result might be roughly comparable to my 6 core result? But only very roughly! To directly compare maybe we'd have to bench 3 full cores of mine vs 3 of yours, so both machines have free cores, with hyperthreading disabled on your pc. Or I try to enable HT on mine, and see a full bench - passmark should have measured performance with turbo boost and HT on with both chips. Then the values could be compared/scaled by the single core passmark scores. You can see the effect of having a newer cpu generation at out single core benching values. Like intel chips can have more power than amd although the amd is running on a much higher clock. Also those cpu scores are probably made for raw computing throughput, and not for latency/responsiveness tasks.

 

Yes, and my machine is reasonably tuned, and with all that renoising activating those tunings I already fear the next electricity bill. It is also that you can really tune linux much more delicate and throughout than windows, and maybe even better than a mac. I also vaguely remember other programs, where I read in the past that the linux versions had more performance, simply because the compiler for linux did a better job and the operating system was less bloated. For realtime performance, the operating system point probably weighs even heavier.

 

I'm looking forward to other linux/3.1 benchmark results now. Did someone already compare differences between renoise 3.0 and 3.1? I think there were changes in the EQ, and I've seen the EQ device is one part of the testing effect chains?


  • Ledger likes this

#47 Ledger

Ledger

    Guruh Motha Fakka Knows More About Renoise Than Taktik

  • Normal Members
  • PipPipPipPipPipPipPipPipPipPipPipPipPipPip
  • 3357 posts
  • Gender:Male

Posted 20 January 2016 - 00:20

Here we go: 

 

Previous Best

 

Xeon 1231 v3

 

HYPER THREADING ENABLED (4 cores = 8 hyperthreaded cores) 

Crimson unplugged and pattern ed out of view,

RAM in proper Dual Channel operation,

High Performance profile in win 7 64bit

AntiVirus Realtime still running

 

-------------------------------------------------
1 core  10ms = 51
8 cores 10ms = 156

1 core  100ms = 60
8 cores 100ms = 227

-------------------------------------------------

 

 

HYPER THREADING DISABLED

Crimson unplugged and pattern ed out of view,

RAM in proper Dual Channel operation,

High Performance profile in win 7 64bit

AntiVirus Realtime still running

 

-------------------------------------------------
1 core  10ms = 51    (=)
3 cores 10ms = 138  (n/a)
4 cores 10ms = 147 (-9)

1 core  100ms = 61   (+1)
3 cores 100ms = 170  (n/a)
4 cores 100ms = 208   (-19)

-------------------------------------------------

 

Edit:   corrected: Wrong previous best posted, so HT disabled loses 8 RPTS on Max cores -- not as much as I would have thought though. 

 

OK I`ve re-run  the Hyperthreading  enabled in same conditions with AV on.  Funnily enough I got best results so far (maybe windows was doing some housekeeping first time or it was full moon or something :badteeth: .. ) 


Edited by Ledger, 20 January 2016 - 17:50.

--> Lua For Beginners <--
--> Lua for newbies <--

My Scripts On Forum

Top Tip!

 

cpu : Xeon 1231 v3, os : Win 7 64bit, audio: Audient iD4
posts as 4tune @ KvR and some other music related sites


#48 somemoron

somemoron

    New Member

  • Normal Members
  • Pip
  • 3 posts
  • Gender:Male
  • Location:London UK
  • Interests:Drums and Renoise

Posted 07 April 2017 - 22:49

Here are my scores:

 

AMD FX-8350 (eight core) @ 4.00Ghz (stock)
Renoise 3.1.0  Win 7(64)

1 core   (10ms): 30

8 cores (10ms): 97

1 core (100ms): 43

8 Cores (100ms): 149

 

I also have a laptop (rather old though)

 

Intel i7 L640@2.13ghz

Renoise 3.1.0  Win 7(64)

 

1 core   (10ms): 21

4 cores (10ms): 42

1 core (100ms): 35

4 Cores (100ms): 64

 

 

Thanks for providing the benchmark Cactoos!


Edited by somemoron, 08 April 2017 - 01:57.