Reason 9.5 & above CPU Stress Tests (2017, two different songfiles included)!

This forum is for discussing Propellerhead's music software. Questions, answers, ideas, and opinions... all apply.
Ad0
Posts: 28
Joined: 13 Jun 2017

18 Nov 2017

Reason 10, Windows 10 x64 professional.

31.3 seconds @ 1024 samples
I got 28 seconds with 128 samples so not really a huge difference.

CPU usage limit set in Reason 9.5 = 95%
Sample rate = 44,100 Hz
Buffer length = 1024 samples

Hardware:
Processor Intel(R) Core(TM) i7-5930K CPU @ 3.50GHz, 3570 Mhz, 6 Core(s), 12 Logical Processor(s)
Motherboard: ASUSTeK COMPUTER INC. X99-DELUXE
Installed Physical Memory (RAM) 16,0 GB
Sound card: RME HDSPe RayDAT

Computer's getting old but I have not experienced any showstoppers in performance for what I do. :)

thhedk
Posts: 1
Joined: 29 Nov 2017

29 Nov 2017

Intel i7 5820k 3,3GHz
MSI X99 Raider
16GB ram PC4-3000MHz
FocusRite Scarlet 6i6 Mk II
Windows 10 64-bit
Reason 10

R9.5 Benchmark44.100HzV2:
Time 35.457

R9.5+2017 SimpleFXChainBenchkmark0:
Time 1:05:086

** Edit, got the last core up running, by choosing Priority High

User avatar
Psuper
Posts: 524
Joined: 29 May 2016

08 Dec 2017

Motherboard: ASRock - Fatal1ty Z370 Gaming-ITX/ac Mini ITX
Processor: Core i7-8700K Vanilla
Memory: 32GB DDR4
Audio Interface: Antelope Discrete 4 (currently on USB).
Windows 10 Vanilla (only optimized to background services, nothing else)
Limit set to 95%

Reason 9.5 2017 COMPLEX RE benchmark song V2
44,100 Hz, 1024 samples
Stopped playing at 50 seconds, bar 26.2
Additional note: all cores over 90% load, max temp was 50c and box was silent, didn't even increase fan speed.

Simple never got above 50% CPU
Reason needs to DAW.viewtopic.php?f=6&t=7504985

User avatar
Kategra
Posts: 262
Joined: 18 Jan 2015

08 Dec 2017

Psuper wrote:
08 Dec 2017
Motherboard: ASRock - Fatal1ty Z370 Gaming-ITX/ac Mini ITX
Processor: Core i7-8700K Vanilla
Memory: 32GB DDR4
Audio Interface: Antelope Discrete 4 (currently on USB).
Windows 10 Vanilla (only optimized to background services, nothing else)
Limit set to 95%

Reason 9.5 2017 COMPLEX RE benchmark song V2
44,100 Hz, 1024 samples
Stopped playing at 50 seconds, bar 26.2
Additional note: all cores over 90% load, max temp was 50c and box was silent, didn't even increase fan speed.

Simple never got above 50% CPU
Amazing performance/$ !!! Thanks for sharing the result!

User avatar
moalla
Posts: 43
Joined: 20 Oct 2017

14 Dec 2017

Now it´s finally done, Kategra my computer is fast as yours, at similia power consumption,
your benchmark demonstrates that the total amount of Megaherz are standing finally over the number of cores

So faster and bigger cores on the processor die, defines in relation with the ram speed
how much power is offered for complex polyphonic workloads.

link latest daw bench 2017
https://techreport.com/review/32642/int ... eviewed/14

In comparison with my latest bench
Ryzen 1700 3800Mhz OC 1,275VCore , Asrock AB350M Pro4, Crucial Ballistix LT 2666<2933Mhz OC CL16, M2 Crucial MX300 525gb, Win7 64 stable, Reason10, Fireface UC, Gtx1050 Ti 6W idle

Image

Complex 9.5 24,2 sec HT on

Simple FX 9.5 2,00 min HT off
1,01 min HT on

max load 88%, highest cpu temperature 58°C 3800, 50°C 3200,

Its unbelievably what a big performance boost pushed now by AMD,
What culminates in X2 processing power at INTEL consumer chips,
offered under 500€, damn but i just bought a computer :roll:

But thats´s okay , at lower load my complete System inkl 1 Display needs 80W-100W power,
the bench shows also 13-20sec for the complex 9.5 Bench up 3000-3600 CPU speed,
at 3800Mhz the power cosumption gets offer 135W+ for the hole System ;)
but if a system isn´t optimized and there are to much extra startup loads ......

It seems that it´s a thing of speed and lager core elements, like faster instruction queues in the cpu, larger cache .....
but when i compare these results with the previous cpu generation, there must be also changes
in the code and virtualization facilitie, however

Intel says that their manufacturing technolog are three years ahead AMD,
so finally I am excited what the Intel Crew from Tel Aviv will offer the users at future,
how developed the core2 duo design :?:
there is no right life in the wrong one [T.W. Adorno]
MAKE REALITY GREAT AGAIN

Thinkpad X220, Ryzen1700, Ballistix 2666, Asrock AB350M, Fireface UC, Audient ID14

Przemyslaw
Posts: 16
Joined: 05 Jan 2018

05 Jan 2018

Motherboard: ASRock Z97M Pro4
Processor: Core i7-4790k @4,7GHz
Memory: 16GB DDR3 (4x4) 2400MHz
Audio Interface: Motu Track 16 - firewire
Windows 7
Limit set to 95%

Reason 9.5 2017 COMPLEX RE benchmark song V2
44,100 Hz, 1024 samples
Stopped playing at 20 seconds

Simple FX 9.5
44,100 Hz, 192 samples
1:20min
44,100 Hz, 1024 samples
1:00min


Upgrade to 7820x soon.

User avatar
Protostar
Posts: 27
Joined: 02 Apr 2017

05 Jan 2018

So i have a few results to show.
Before today i was running an i7 6700K @ 4.5GHz (4 core 8 thread), with an Asus maximus viii Hero & 16gb of 3000MHz DDR4.
Also i am using a Focusrite Scarlett 2i4 on both configs, at 44.1k and 884 samples, looking to swap the interface out for something better soon too.

Here are my results on that config:
Complex file 16.8 secs
Simple Chain 1 min 21 secs

I upgraded to an i9 7900x, with an Asus TUF X299 Mark 1 mobo, also getting 32gb of 3000MHz Quad channel Gskill ram next week, but for these tests right now i am only running in dual channel with the same 16Gb of ram, will update when i get new ram in, though i suspect it wont affect times much.

Complex file @ stock 3.3GHz 45.7 secs
Simple chain @ stock 3.3GHz 1 min 15 secs

And now overclocked to 4.3GHz on all cores @ 1.05V with max stress test temps of 80c on a 240mm AIO cooler:

Complex file @ 4.3Ghz 48.5 secs
Simple chain @ 4.3GHz 1 min 16 secs


And here are some geekbench scores of all 3 configs:
6700k @ 4.5GHz
7900X @ 3.3GHz
7900X @ 4.3GHz


Now i will probably be able to bump these numbers even higher with a soundcard that has a better driver/converter in it, like an RME interface or antelope, but for now its still a huge improvement over the 6700k and im very happy so far!
I knew id be getting worse singel core performance as the i7 was a gaming chip and still is extremely good at single threaded workloads, but im ok with trading off a bit of single core power for a huge increase in multithreading.

Now lets hope propheads do a lot of optimisation work on reason & its VST implementation soon :D

User avatar
syncanonymous
Posts: 338
Joined: 16 Mar 2015
Location: UK and France

11 Jan 2018

Psuper wrote:
08 Dec 2017
Motherboard: ASRock - Fatal1ty Z370 Gaming-ITX/ac Mini ITX
Processor: Core i7-8700K Vanilla
Memory: 32GB DDR4
Audio Interface: Antelope Discrete 4 (currently on USB).
Windows 10 Vanilla (only optimized to background services, nothing else)
Limit set to 95%

Reason 9.5 2017 COMPLEX RE benchmark song V2
44,100 Hz, 1024 samples
Stopped playing at 50 seconds, bar 26.2
Additional note: all cores over 90% load, max temp was 50c and box was silent, didn't even increase fan speed.

Simple never got above 50% CPU
All research I been doing this past week on a new workstation is pointing at this exact same CPU and mobo with DDR4 that you have here, Psuper.

User avatar
Protostar
Posts: 27
Joined: 02 Apr 2017

14 Jan 2018

syncanonymous wrote:
11 Jan 2018
Psuper wrote:
08 Dec 2017
Motherboard: ASRock - Fatal1ty Z370 Gaming-ITX/ac Mini ITX
Processor: Core i7-8700K Vanilla
Memory: 32GB DDR4
Audio Interface: Antelope Discrete 4 (currently on USB).
Windows 10 Vanilla (only optimized to background services, nothing else)
Limit set to 95%

Reason 9.5 2017 COMPLEX RE benchmark song V2
44,100 Hz, 1024 samples
Stopped playing at 50 seconds, bar 26.2
Additional note: all cores over 90% load, max temp was 50c and box was silent, didn't even increase fan speed.

Simple never got above 50% CPU
All research I been doing this past week on a new workstation is pointing at this exact same CPU and mobo with DDR4 that you have here, Psuper.
Bear in mind you wont get the same results as him unless you have the soundcard to back it up :) Friend of mine has the 8700k and can only get around 25 secs in the benchmark, not 50

I get 50 with an i9 and a shitty soundcard, so yeah you might not get same performance unless you have a really nice interface

Przemyslaw
Posts: 16
Joined: 05 Jan 2018

14 Jan 2018

Hi Everyone !

I make new setup.
But before that I present the best score on the old setup: i7 4790k OC 4.7GHz, RAM 16GB DDR3 2400MHz - 24,265 sec in Complex benchmark song. Operation system -windows 7 utimate was installed in 2014, benchmark made week ago.
i7 4790k


New Setup:
Motherboard: ASUS TUF X299 Mark 1
Processor: i7 7820x (stock - 3,6GHz with Turbo Boost2 - 4,3GHz & TB3 4,5GHz)
Memory: 16GB DDR4 (4x4) 2800MHz CL15
Audio Interface: Motu Track 16 - usb (set to 1024 samples)
Windows 10 Home x64
Reason 9.5
Limit set to 95%

Because on the new setup I was not satisfied about performance (around 40sec in complex bench) especially compared to the great result of Psuper on his 8700k and Protostar 7900x I did a lot of work to optimize the windows system and set the bios. In addition, I introduced a delicate overlocking. Of course had in mind great xeon results too !

Best Score setup:
Cpu 7820x OC 4,6GHz (fixed)
Ram 3200MHz CL15
In Complex benchmark song I've got 56,517 sec
In the Passmark bench I've got 20427points - better than the past king of desktop cpu - i7 6950x
7820x OC 4,6Ghz & RAM 3200MHz CL15
Passmark result

After all, I decided to set the computer for daily work in the following specifications:
Cpu 7820x OC 4,3GHz (fixed)
Ram 3000MHz CL15
In Complex benchmark song I've got 52,315 sec
7820x Oc 4,3GHz & RAM 3000MHz CL15
Simple benchmark song around 1:15 min with HT on and around 2:00 min with HT off


In case someone need this specific info in the future I observed that RAM has great input to the overall performance in Reason workflow. Greater than the cpu OC. I read that this is common on a Skylake chipset. In detail - cpu overlocked from 4,3GHz (fixed) to 4,6Ghz (fixed) improved Reason performance in benchmark complex song for about 1-2 extra seconds. Overlocked RAM from 2800MHz CL15 to 3200MHz CL15 improved Reason performance in benchmark complex song for about 8 extra seconds.
Someone from Reason community could make test and confirm that observation.

User avatar
Psuper
Posts: 524
Joined: 29 May 2016

14 Jan 2018

Protostar wrote:
14 Jan 2018
syncanonymous wrote:
11 Jan 2018


All research I been doing this past week on a new workstation is pointing at this exact same CPU and mobo with DDR4 that you have here, Psuper.
Bear in mind you wont get the same results as him unless you have the soundcard to back it up :) Friend of mine has the 8700k and can only get around 25 secs in the benchmark, not 50

I get 50 with an i9 and a shitty soundcard, so yeah you might not get same performance unless you have a really nice interface
It doesn't make much of a difference what card you use.

I ran the V2 again with "DX Primary Sound Driver" at 44.1 / 1024 samples and got 47 seconds, bar 24.3.

Keep in mind while everyone likes raw performance, I focused this build on cool and quiet (and offline once I decide if I'm keeping the Discreet 4 or not), hence some of the unconventional parts and way it's built. At max load I hit 46c and total silence, no case fans running, cpu fan stayed at lowest rpm.
Reason needs to DAW.viewtopic.php?f=6&t=7504985

User avatar
Protostar
Posts: 27
Joined: 02 Apr 2017

15 Jan 2018

Psuper wrote:
14 Jan 2018
Protostar wrote:
14 Jan 2018


Bear in mind you wont get the same results as him unless you have the soundcard to back it up :) Friend of mine has the 8700k and can only get around 25 secs in the benchmark, not 50

I get 50 with an i9 and a shitty soundcard, so yeah you might not get same performance unless you have a really nice interface
It doesn't make much of a difference what card you use.

I ran the V2 again with "DX Primary Sound Driver" at 44.1 / 1024 samples and got 47 seconds, bar 24.3.

Keep in mind while everyone likes raw performance, I focused this build on cool and quiet (and offline once I decide if I'm keeping the Discreet 4 or not), hence some of the unconventional parts and way it's built. At max load I hit 46c and total silence, no case fans running, cpu fan stayed at lowest rpm.
Those results dont make much sense to me. My friend has the exact same chip as you and can only get around 26 secs on dx primary sound driver, and youre not even overclocked on your chip.

Is there something special you're doing in windows to get this result? Or are you prepping the song file in any way by minimising devices etc?

Edit: These are some results on my i9 with optimising for background services + performance

Performance & Background: https://i.imgur.com/aSFkDdw.png
Appearance & Background: https://i.imgur.com/qB6KtTV.png
Appearance & Programs: https://i.imgur.com/qB6KtTV.png

Not much of a difference really

User avatar
Psuper
Posts: 524
Joined: 29 May 2016

15 Jan 2018

Protostar wrote:
15 Jan 2018
Psuper wrote:
14 Jan 2018


It doesn't make much of a difference what card you use.

I ran the V2 again with "DX Primary Sound Driver" at 44.1 / 1024 samples and got 47 seconds, bar 24.3.

Keep in mind while everyone likes raw performance, I focused this build on cool and quiet (and offline once I decide if I'm keeping the Discreet 4 or not), hence some of the unconventional parts and way it's built. At max load I hit 46c and total silence, no case fans running, cpu fan stayed at lowest rpm.
Those results dont make much sense to me. My friend has the exact same chip as you and can only get around 26 secs on dx primary sound driver, and youre not even overclocked on your chip.

Is there something special you're doing in windows to get this result? Or are you prepping the song file in any way by minimising devices etc?

Edit: These are some results on my i9 with optimising for background services + performance

Performance & Background: https://i.imgur.com/aSFkDdw.png
Appearance & Background: https://i.imgur.com/qB6KtTV.png
Appearance & Programs: https://i.imgur.com/qB6KtTV.png

Not much of a difference really
As I mentioned I haven't even begun optimizing my system software or hardware - just raw choice components at the moment vanilla everything. Either he has HT off or there's something wrong with his hardware setup. Have him run the benchmark software on http://www.userbenchmark.com/ so he can see where he sits with others of similar hardware.
Reason needs to DAW.viewtopic.php?f=6&t=7504985

User avatar
syncanonymous
Posts: 338
Joined: 16 Mar 2015
Location: UK and France

15 Jan 2018

Psuper wrote:
15 Jan 2018
Protostar wrote:
15 Jan 2018


Those results dont make much sense to me. My friend has the exact same chip as you and can only get around 26 secs on dx primary sound driver, and youre not even overclocked on your chip.

Is there something special you're doing in windows to get this result? Or are you prepping the song file in any way by minimising devices etc?

Edit: These are some results on my i9 with optimising for background services + performance

Performance & Background: https://i.imgur.com/aSFkDdw.png
Appearance & Background: https://i.imgur.com/qB6KtTV.png
Appearance & Programs: https://i.imgur.com/qB6KtTV.png

Not much of a difference really
As I mentioned I haven't even begun optimizing my system software or hardware - just raw choice components at the moment vanilla everything. Either he has HT off or there's something wrong with his hardware setup. Have him run the benchmark software on http://www.userbenchmark.com/ so he can see where he sits with others of similar hardware.
“same chip”...meaning 8700K far as I understand...what about other components...mobo and RAM...wifi adapter on or off? RAM speed? there are a lot of factors, innit? my plan is aiming at silent as poss and low temps; air cooling

User avatar
Protostar
Posts: 27
Joined: 02 Apr 2017

16 Jan 2018

syncanonymous wrote:
15 Jan 2018
Psuper wrote:
15 Jan 2018


As I mentioned I haven't even begun optimizing my system software or hardware - just raw choice components at the moment vanilla everything. Either he has HT off or there's something wrong with his hardware setup. Have him run the benchmark software on http://www.userbenchmark.com/ so he can see where he sits with others of similar hardware.
“same chip”...meaning 8700K far as I understand...what about other components...mobo and RAM...wifi adapter on or off? RAM speed? there are a lot of factors, innit? my plan is aiming at silent as poss and low temps; air cooling
My friend didn't tell me he was only running 1 stick of 16gb ddr4, so its only running single channel memory atm. Made him order 2 new sticks to run in dual channel config for double bandwidth, so will update with those results.

But just so you know, most motherboards wont impact your benchmarks whatsoever really, and same goes for ram speed (in music at least), even a difference in CAS latency wont do much, its moreso about the bandwidth, same goes for having a wifi adapter on/off.

Only real speed impacts will be cpu, ram (within reason, as long as youre running it in dual channel or quad channel if your chip can do quad channel like my i9), and your GPU (to take the load off of your CPU)

User avatar
Psuper
Posts: 524
Joined: 29 May 2016

16 Jan 2018

Protostar wrote:
16 Jan 2018

My friend didn't tell me he was only running 1 stick of 16gb ddr4, so its only running single channel memory atm. Made him order 2 new sticks to run in dual channel config for double bandwidth, so will update with those results.
Yikes... talk about a overarching bottleneck for a new system!
Reason needs to DAW.viewtopic.php?f=6&t=7504985

User avatar
syncanonymous
Posts: 338
Joined: 16 Mar 2015
Location: UK and France

17 Jan 2018

Psuper wrote:
16 Jan 2018
Protostar wrote:
16 Jan 2018

My friend didn't tell me he was only running 1 stick of 16gb ddr4, so its only running single channel memory atm. Made him order 2 new sticks to run in dual channel config for double bandwidth, so will update with those results.
Yikes... talk about a overarching bottleneck for a new system!
that ‘splains it!

User avatar
Protostar
Posts: 27
Joined: 02 Apr 2017

17 Jan 2018

Psuper wrote:
16 Jan 2018
Protostar wrote:
16 Jan 2018

My friend didn't tell me he was only running 1 stick of 16gb ddr4, so its only running single channel memory atm. Made him order 2 new sticks to run in dual channel config for double bandwidth, so will update with those results.
Yikes... talk about a overarching bottleneck for a new system!
Yup, with dual channel RAM he gets 46 secs now lmao, everything adds up now haha!

Very nice CPU for cost!

Been considering making a cross DAW VST benchmark to compare Reason vs Fl Studio etc in VST performance too, will post if i end up making it!
Last edited by Protostar on 18 Jan 2018, edited 1 time in total.

househoppin09
Posts: 510
Joined: 03 Aug 2016

17 Jan 2018

Protostar wrote:
17 Jan 2018
Been considering making a cross DAW VST benchmark to compare Reason vs Fl Studio etc in VST performance too, will post if i end up making it!
I'd really love to see this.

User avatar
Protostar
Posts: 27
Joined: 02 Apr 2017

18 Jan 2018

househoppin09 wrote:
17 Jan 2018
Protostar wrote:
17 Jan 2018
Been considering making a cross DAW VST benchmark to compare Reason vs Fl Studio etc in VST performance too, will post if i end up making it!
I'd really love to see this.
Me and a friend have made it, and the results are quite saddening haha. Posting soon, ill link thread here when its up :)

Edit: here it is
Last edited by Protostar on 18 Jan 2018, edited 1 time in total.

User avatar
SH1audio
Posts: 4
Joined: 18 Jan 2018

18 Jan 2018

System:
i7 8700K OC @ 5 GHZ
32 GB Crucial Ballistix Sport DDR4 2400 MHZ
ASUS Prime Z370 Series Mobo
Presonus Audiobox iTwo
DX Primary Sound Driver @4096
R 9.5 Bench.PNG
R 9.5 Bench.PNG (291.07 KiB) Viewed 1664 times
It got quite choppy at the end but that's due to the standard driver methinks.

User avatar
bsp
Posts: 214
Joined: 18 Jan 2015

21 Jan 2018

config:
- i9 7980xe @2600MHz (no overclocking)
- 32gb RAM (@1333MHz, quad channel (4x8gb))
- Focusrite Scarlett 18i20 USB2 audio interface
- Windows 10 pro 64bit

time until dropouts and "Computer too slow" message:

complex:
1:16m 64 samples (dropouts starting at 0:30)
1:20m 128 samples (dropouts starting at 0:30)
1:20m 512 samples (dropouts starting at 0:35)
1:21m 1024 samples (dropouts starting at 1:20)

fxchain:
2:06m 64 samples (dropouts starting at ~0:19, lots of them at ~0:28)
1:49m 128 samples (dropouts starting at ~0:22, lots of them at ~0:35)
1:33m 256 samples (dropouts starting at 0:48, lots of them at ~1:00)
1:18m 512 samples (no dropouts)
1:18m 1024 samples (no dropouts)

notes:
- max. 181W total measured at wall socket
- max. CPU Temperate 45°C
- max. 98W for CPU (according to IA Suite 3)
- "balanced" OS power plan
- OS optimized for background services
- bios: cstates off (except for c1..)
- bios: hyperthreading off
- bios: turbomode off
- sample rate set to 44.1kHz
- Reason CPU usage limit set to 95%


I've only had this PC for a few days now and it has massive overclocking potential.
There's one user here who has an i9 OC'd to 5GHz, would be interesting to see the benchmark results on that machine (and the power consumption ;)).

One thing that is really astonishing about this CPU is how efficient it is.
In a regular "office"-like usage scenario, the whole system consumes ~125W (~45W for the CPU, ~36W with all cstates enabled, 7W for the whole system when it's sleeping, ~38°C CPU temperature under normal load).

When compiling large projects (I am also using this for software development), power consumption goes up to ~140W (~42°C CPU temperature), and the build times are actually 4 times shorter than on my previous machine.

When it comes to audio production, I think you're currently better off with a fast 8 (or maybe 10) core CPU, since the software (tested with Reason+Reaper) is apparently not able to fully utilize all cores properly -- also see my other comments in my post in the Reason 8.1 benchmark thread.

Last but not least, these benchmarking "games" should always be taken with a big grain of salt:
As you can see, audio dropouts are usually starting long before the "too slow" message, and at least as far as I am concerned, I want near-realtime performance since I usually arrange+modulate things "live" (using my custom made sequencer + Reason as a soundgenerator+tape recorder).

After seeing the somewhat disappointing multi-core utilization in both Reason and Reaper, I'm very eager to see what a VST host that's been specifically optimized for many-core CPUs will be able to accomplish (currently working on it and have so far tested it with more than 400 different plugins). I _hope_ that it will be able to run 14..16 instances of Diva but we'll see ;)

Don't get me wrong: Reason is quite capable of multi-core rendering, it only gets problematic when single tracks/combinators exceed ~4 CPU bars.
It's quite possible to run a large amount of moderately heavy synths like Respire, eXpanse, Diva, Bazille, etc on this machine (tested with up to 16 instances of Respire).
My old PC (i7 4770k) could not run more than 3..

Przemyslaw
Posts: 16
Joined: 05 Jan 2018

21 Jan 2018

Great results bsp.

Dropouts depend on the whole system configuration, sound card, os optimization, peripherals, connections, applications in the background.
For example in complex reason benchmark song with 512 samples of latency I have dropouts start at 32sec (hyperthreading off in Reason), and start at 36sec (HT on).
Cpu i7 7820x, ram 3000mhz, Motu Track 16 usb, system optimized, cpu fixed.
I think You can easly beat your results with that amount of computing power of your cpu after some optimization and modify configuration.

You Have 2666MHz Ram. Set it on 3200MHz (You said in other post to have 3200MHz native ram speed) and check the performance increase in reason complex if You want.

User avatar
bsp
Posts: 214
Joined: 18 Jan 2015

21 Jan 2018

I set the RAM clock to "3200Mhz" in the BIOS (i.e. 2x1600MHz) and repeated the complex test @1024 samples.
The result was exactly the same as with the 2666 setting ("too slow" after 1:21min).

Then I enabled the CPU "turbo mode" (overclocks all 18 cores up to 3400MHz).

Now the complex test finished without dropouts (@512 and 1024 samples) :-) (max CPU temp. 46°, ~190W total power consumption).
@256 samples it also plays until the end but crackles a bit.

With turbo mode enabled, the fxchain test stops playing @1:36 (@1024 samples, no dropouts).
At a 256 sample buffer size, dropouts start at ~0:55 and the "too slow" message appears at 2:10, @64 samples at 2:44.

Przemyslaw
Posts: 16
Joined: 05 Jan 2018

21 Jan 2018

Mega ! So its time to make complex V3 Kategra :)

Well, my "ram theory" isn't correct in your case.

According to your results, many users have had better results in fxchain benchmark with HT off in Reason.
After a few days with my newly assembled computer, I have the HT option turned off, recently I noticed that the project with many respire, expanse, plus audio tracks, vst's and other rack extensions has more stability without hyperthreading. But it seems that it depends on the specifics of each song.

Post Reply
  • Information
  • Who is online

    Users browsing this forum: No registered users and 8 guests