Guys what is the best cpu for reason?

This forum is for discussing Propellerhead's music software. Questions, answers, ideas, and opinions... all apply.
reggie1979
Posts: 1176
Joined: 11 Apr 2019

Post 13 Aug 2019

EnochLight wrote:
13 Jul 2019
reggie1979 wrote:
12 Jul 2019
I just hope it works at lower latencies. Seems that was a potential problem early on.
To be clear, all of the Ryzen/Zen generation CPU’s work 100% fine at lower latencies. It’s just that Intel chips will offer slightly more plugins at low latencies (64 samples, etc). But the difference isn’t massive IIRC...
Still confused. Are you saying that you'll get less at an equivalent clock speed? If so I guess then that is what you pay for since Intel is generally far more expensive.

I'm glad I'm not going through this right now, though one of those Thread ripper chips must be laughable at how many plugs you can run!

sled9e
Posts: 5
Joined: 02 Jun 2016

Post 13 Aug 2019

In my case i think the extra cores come in handy when rendering 3d images/motions.graphical stuff

reggie1979
Posts: 1176
Joined: 11 Apr 2019

Post 13 Aug 2019

Well, if money was not an issue, I'd just want bragging rights :D

But, I don't have the money and simply do not need anything other than what I have.

RealReasonHead
Posts: 33
Joined: 21 Jun 2019

Post 13 Aug 2019

EnochLight wrote:
13 Jul 2019
To be clear, all of the Ryzen/Zen generation CPU’s work 100% fine at lower latencies. It’s just that Intel chips will offer slightly more plugins at low latencies (64 samples, etc). But the difference isn’t massive IIRC...
Not anymore! The 3900X is at least on par and in some tests even ahead of the 9900K.
zagrad wrote:
09 Aug 2019
You seem to mention Intel Core i9-9900K @ 3.60GHz a lot if I've seen correctly, so I made a comparison to that one.
https://www.cpubenchmark.net/compare/AM ... 3494vs3334
cpubenchmark might not be the best website to compare CPUs for your purpose. Some tech youtubers (Steve from Gamers Nexus in particular) even say it's absolute garbage at comparing anything. CPU performance depends on a lot of variables (but mainly paired RAM and application). The single core performance of the 3000 chips is in some cases even better than the Intel's 9th gen (Cinebench scores, for example). A use case specific benchmark like the dawbench link I've linked above makes for a better comparison, imo.

User avatar
EnochLight
Posts: 5912
Joined: 17 Jan 2015
Location: your mom

Post 14 Aug 2019

RealReasonHead wrote:
13 Aug 2019
EnochLight wrote:
13 Jul 2019
To be clear, all of the Ryzen/Zen generation CPU’s work 100% fine at lower latencies. It’s just that Intel chips will offer slightly more plugins at low latencies (64 samples, etc). But the difference isn’t massive IIRC...
Not anymore! The 3900X is at least on par and in some tests even ahead of the 9900K.
Except that in the link you posted, the 3900X still suffers at low latencies.

The 3900X has the noted performance latency still, although it seems to vary between applications and we don’t see that occurring with either the Reaper or Cubase test on the Intel side. I wouldn’t normally be happy with seeing anything drop out at 70% or 80% load but there is certainly an argument that it still offers reasonable value as even then it exceeds the 9900K which is currently sat around the same price.
RealReasonHead wrote:
13 Aug 2019
zagrad wrote:
09 Aug 2019
You seem to mention Intel Core i9-9900K @ 3.60GHz a lot if I've seen correctly, so I made a comparison to that one.
https://www.cpubenchmark.net/compare/AM ... 3494vs3334
cpubenchmark might not be the best website to compare CPUs for your purpose. Some tech youtubers (Steve from Gamers Nexus in particular) even say it's absolute garbage at comparing anything. CPU performance depends on a lot of variables (but mainly paired RAM and application). The single core performance of the 3000 chips is in some cases even better than the Intel's 9th gen (Cinebench scores, for example). A use case specific benchmark like the dawbench link I've linked above makes for a better comparison, imo.
Until I am thoroughly familiar with some random YouTuber’s testing methods, I’d still default to CPUbenchmark’s results. The results are industry tested and sample based from various user builds.

That said, it looks like the 3900X is still outperforming the Intel 9900K, which is pretty bad ass.
Win 10 | Reason 11 |  Studio One 4.5 | i7 3770k @ 3.5 Ghz | 16 GB RAM | Reactor 1TB SSD | RME Babyface Pro | Nektar Panorama P-4 | Akai MPC Live | Roland System 8 and System 1 | Roland TR-8 with 7x7 Expansion | Roland TB-3 | Roland VT-4 | Roland MX-1

zagrad
Posts: 4
Joined: 09 Aug 2019

Post 14 Aug 2019

The 3900X has the noted performance latency still, although it seems to vary between applications and we don’t see that occurring with either the Reaper or Cubase test on the Intel side. I wouldn’t normally be happy with seeing anything drop out at 70% or 80% load but there is certainly an argument that it still offers reasonable value as even then it exceeds the 9900K which is currently sat around the same price.
Ok, so about that performance latency. I think I'm still not sure what that means or how it would affect the application.
This may be a stupid question, but what exactly happens in Reason when the performance latency problem kicks in? Or is it something that is there all the time? When will I notice it? Only when there are tons of tracks and tons of synths?

Oh and thanks for all the info already guys :)

zagrad
Posts: 4
Joined: 09 Aug 2019

Post 14 Aug 2019

RealReasonHead wrote:
13 Aug 2019
Not anymore! The 3900X is at least on par and in some tests even ahead of the 9900K.
Interesting. The writer of the article though also states this:
My only reservation at this time is compatibility with third-party hardware and mainly interfaces. We saw some compatibility issues with Ryzen 1 & 2 with some PCIe sound cards and some USB based interfaces. ASMedia have a bit of a poor rep on Intel board where they’ve provided their third party USB3 solutions as audio devices don’t tend to play too well on them. We saw similar incidents with the implementation they packaged for the Ryzen board on generation one and thankfully it was less common on Ryzen gen 2.

Ryzen Gen 3 has an AMD designed USB implementation but built around an ASMedia package and at this point, I’ve little idea how it will hold up with all the device we have available. I was testing using a Babyface Pro this time around, so that’s validated, but I would certainly check with user groups for your key devices for any compatibility issues prior to buying.
That's something I haven't even thought about and may be an even bigger problem. If the CPU performs well, but doesn't work well with a certain audio interface it's still useless right?
Does anyone know how big of a problem that is?

I was leaning towards the AMD chip, but this makes me lean towards Intel a lot.

RealReasonHead
Posts: 33
Joined: 21 Jun 2019

Post 14 Aug 2019

EnochLight wrote:
14 Aug 2019
Except that in the link you posted, the 3900X still suffers at low latencies.
Are we looking at the same charts? In the first picture (always comparing it the similarly priced 9900K), the 3900X outperforms it by a big margin, 117 vs 91. In the second chart, it is still ahead, 400 vs 380. Always at a buffer size of 64 samples.
EnochLight wrote:
14 Aug 2019
Until I am thoroughly familiar with some random YouTuber’s testing methods, I’d still default to CPUbenchmark’s results. The results are industry tested and sample based from various user builds.
Steve is well respected in the tech community and provides very thorough and well-controlled tests - the best tests I've seen on Youtube anyways. He knows his stuff. I can link to the video where he quickly addressed his issue with CPUbenchmark, if you're interested. The small "scandal" with CPUbenchmark definitely applies to audio, imho. Multicore performance is pretty important, if you don't have very few tracks.

RealReasonHead
Posts: 33
Joined: 21 Jun 2019

Post 14 Aug 2019

zagrad wrote:
14 Aug 2019
Ok, so about that performance latency. I think I'm still not sure what that means or how it would affect the application.
This may be a stupid question, but what exactly happens in Reason when the performance latency problem kicks in? Or is it something that is there all the time? When will I notice it? Only when there are tons of tracks and tons of synths?
If I understand their argument correctly it's about how much of the CPU can be utilized until you get buffer underruns. This is taken into account in the link I've posted. Realtime applications (like audio) have different requirements for the CPU than, say, rendering. Ryzen wasn't that great of a choice for those tasks until Zen 2.

reggie1979
Posts: 1176
Joined: 11 Apr 2019

Post 14 Aug 2019

zagrad wrote:
14 Aug 2019
RealReasonHead wrote:
13 Aug 2019
Not anymore! The 3900X is at least on par and in some tests even ahead of the 9900K.
Interesting. The writer of the article though also states this:
My only reservation at this time is compatibility with third-party hardware and mainly interfaces. We saw some compatibility issues with Ryzen 1 & 2 with some PCIe sound cards and some USB based interfaces. ASMedia have a bit of a poor rep on Intel board where they’ve provided their third party USB3 solutions as audio devices don’t tend to play too well on them. We saw similar incidents with the implementation they packaged for the Ryzen board on generation one and thankfully it was less common on Ryzen gen 2.

Ryzen Gen 3 has an AMD designed USB implementation but built around an ASMedia package and at this point, I’ve little idea how it will hold up with all the device we have available. I was testing using a Babyface Pro this time around, so that’s validated, but I would certainly check with user groups for your key devices for any compatibility issues prior to buying.
That's something I haven't even thought about and may be an even bigger problem. If the CPU performs well, but doesn't work well with a certain audio interface it's still useless right?
Does anyone know how big of a problem that is?

I was leaning towards the AMD chip, but this makes me lean towards Intel a lot.
This is why I brought it up. I've heard about things like this and I'd hate to build a new machine and have this happen. Intel has it's issues but for me mine works.

User avatar
EnochLight
Posts: 5912
Joined: 17 Jan 2015
Location: your mom

Post 15 Aug 2019

zagrad wrote:
14 Aug 2019
RealReasonHead wrote:
13 Aug 2019
Not anymore! The 3900X is at least on par and in some tests even ahead of the 9900K.
Interesting. The writer of the article though also states this:
My only reservation at this time is compatibility with third-party hardware and mainly interfaces. We saw some compatibility issues with Ryzen 1 & 2 with some PCIe sound cards and some USB based interfaces. ASMedia have a bit of a poor rep on Intel board where they’ve provided their third party USB3 solutions as audio devices don’t tend to play too well on them. We saw similar incidents with the implementation they packaged for the Ryzen board on generation one and thankfully it was less common on Ryzen gen 2.

Ryzen Gen 3 has an AMD designed USB implementation but built around an ASMedia package and at this point, I’ve little idea how it will hold up with all the device we have available. I was testing using a Babyface Pro this time around, so that’s validated, but I would certainly check with user groups for your key devices for any compatibility issues prior to buying.
That's something I haven't even thought about and may be an even bigger problem. If the CPU performs well, but doesn't work well with a certain audio interface it's still useless right?
Does anyone know how big of a problem that is?

I was leaning towards the AMD chip, but this makes me lean towards Intel a lot.
That’s a great point, and very much worthy of consideration. Every single AMD platform I ever had suffered from some hardware incompatibility at some point, though it’s been many years since I’ve built an AMD system - they have gotten better.

Anyway, it’s great ScanPro Audio did his testing using an RME Babyface Pro - because that’s my interface. :)
Win 10 | Reason 11 |  Studio One 4.5 | i7 3770k @ 3.5 Ghz | 16 GB RAM | Reactor 1TB SSD | RME Babyface Pro | Nektar Panorama P-4 | Akai MPC Live | Roland System 8 and System 1 | Roland TR-8 with 7x7 Expansion | Roland TB-3 | Roland VT-4 | Roland MX-1

User avatar
EnochLight
Posts: 5912
Joined: 17 Jan 2015
Location: your mom

Post 15 Aug 2019

RealReasonHead wrote:
14 Aug 2019
zagrad wrote:
14 Aug 2019
Ok, so about that performance latency. I think I'm still not sure what that means or how it would affect the application.
This may be a stupid question, but what exactly happens in Reason when the performance latency problem kicks in? Or is it something that is there all the time? When will I notice it? Only when there are tons of tracks and tons of synths?
If I understand their argument correctly it's about how much of the CPU can be utilized until you get buffer underruns. This is taken into account in the link I've posted. Realtime applications (like audio) have different requirements for the CPU than, say, rendering. Ryzen wasn't that great of a choice for those tasks until Zen 2.
Was Zen2 used on all Ryzen 2 & Threadripper 2? Because ScanPro Audio found that the dye latency issue was present on those as well.
Win 10 | Reason 11 |  Studio One 4.5 | i7 3770k @ 3.5 Ghz | 16 GB RAM | Reactor 1TB SSD | RME Babyface Pro | Nektar Panorama P-4 | Akai MPC Live | Roland System 8 and System 1 | Roland TR-8 with 7x7 Expansion | Roland TB-3 | Roland VT-4 | Roland MX-1

User avatar
EnochLight
Posts: 5912
Joined: 17 Jan 2015
Location: your mom

Post 15 Aug 2019

RealReasonHead wrote:
14 Aug 2019
EnochLight wrote:
14 Aug 2019
Except that in the link you posted, the 3900X still suffers at low latencies.
Are we looking at the same charts? In the first picture (always comparing it the similarly priced 9900K), the 3900X outperforms it by a big margin, 117 vs 91. In the second chart, it is still ahead, 400 vs 380. Always at a buffer size of 64 samples.
EnochLight wrote:
14 Aug 2019
Until I am thoroughly familiar with some random YouTuber’s testing methods, I’d still default to CPUbenchmark’s results. The results are industry tested and sample based from various user builds.
Steve is well respected in the tech community and provides very thorough and well-controlled tests - the best tests I've seen on Youtube anyways. He knows his stuff. I can link to the video where he quickly addressed his issue with CPUbenchmark, if you're interested. The small "scandal" with CPUbenchmark definitely applies to audio, imho. Multicore performance is pretty important, if you don't have very few tracks.
Yes - I acknowledged the performance of the new Ryzen was awesome at the end of my post. But are we reading the same article? As the article in your link acknowledged that the 3900X still suffers from latency issues despite that.
Win 10 | Reason 11 |  Studio One 4.5 | i7 3770k @ 3.5 Ghz | 16 GB RAM | Reactor 1TB SSD | RME Babyface Pro | Nektar Panorama P-4 | Akai MPC Live | Roland System 8 and System 1 | Roland TR-8 with 7x7 Expansion | Roland TB-3 | Roland VT-4 | Roland MX-1

RealReasonHead
Posts: 33
Joined: 21 Jun 2019

Post 15 Aug 2019

EnochLight wrote:
15 Aug 2019
But are we reading the same article? As the article in your link acknowledged that the 3900X still suffers from latency issues despite that.
Though at 64 samples the CPU usage is only 70% it still outperforms the 9900K. We are only concerned about pure performance and not how much of the CPU is being utilized, right? Sure, it would be great, if there was a 100% CPU usage at 64 samples but if it still outperforms the similarly priced 9900K it doesn't really matter.
EnochLight wrote:
15 Aug 2019
Was Zen2 used on all Ryzen 2 & Threadripper 2? Because ScanPro Audio found that the dye latency issue was present on those as well.
Nope, Ryzen 2000 and Threadripper 2 are Zen+. IIRC, with Zen 2 they have worked a lot on reducing latency inside the CPU (between the chiplets, this has nothing to do with the buffer sizes) which I assume is the reason, they perform better than the previous generation.

reggie1979
Posts: 1176
Joined: 11 Apr 2019

Post 15 Aug 2019

See, You have to understand what I'm looking for if I'm gonna pony ANYTHING up for what I have. I have a 4770k (not overclocked) that runs Reason, fairly well. It's not ideal, but it doesn't suck.

What I want (me me me) is holy crap jesus christ mother of GOD over that. But currently I run my RME at a 64 buffer and it's fine. I want CRUSHING DEATH and lower latencies with historicly low CPU use. I want 32 buffer and zero bars!!!!!!!!!!! And crazy good performance.

Or, it's not happening.

reggie1979
Posts: 1176
Joined: 11 Apr 2019

Post 15 Aug 2019

BTW, should be reasonable with a thread ripper.

fraz
Posts: 4
Joined: 29 Jul 2018

Post 12 Oct 2019

Recently was working on a system, Asus Z9PE-D8 WS & 2 X E5 2687 W V2 - base clock OC 108 scoring 26,750 multi - and 2,330 single threaded on the Passmark software - not bad

sled9e
Posts: 5
Joined: 02 Jun 2016

Post 24 Oct 2019

Finished mine AMD 3900x build, on An MSI prestige Creation x570 with 64gb (love a lot of USB ports) havent got around to kind of stressing it but so far so good.might do some testing /stressing it later on.

  • Information
  • Who is online

    Users browsing this forum: Chizmata, PGR, TritoneAddiction and 4 guests