Benchmark Test between 10.2 and the upcoming 10.3

[EnochLight]

Looks like there will be little reasons to stay on 9.5 or lower. Good job PH.

Oh make no mistake, there will always be those who think there's a ton of reasons to stay on 9.5 or lower. Trying to convince them otherwise is an exercise in futility.

[antic604]

[re-post from beta subforum]

Same settings, same loop, the project uses almost all native devices, several REs (Parsec, The Legend, FM4, Synchronous, Polar, Fat Space, GQ-7, Distillery) and Phoscyon VST.

Guess which one is which?

Not only is v10.3 performing better (even with -5% lower CPU clock!), but it's also much more stable

[bxbrkrz]

Looks like there will be little reasons to stay on 9.5 or lower. Good job PH.

Oh make no mistake, there will always be those who think there's a ton of reasons to stay on 9.5 or lower. Trying to convince them otherwise is an exercise in futility.

While expecting 110% support for it forever.

[boingy]

Oh make no mistake, there will always be those who think there's a ton of reasons to stay on 9.5 or lower. Trying to convince them otherwise is an exercise in futility.

While expecting 110% support for it forever.

I'm a 9.5-er. Not upgrading was purely a cost vs gains decision for me. R10 had hardly anything to interest me. Yet another new synth I don't need, yet another piano I don't need, a few other average instruments. The only items of interest were Synchronous and the smattering of sequencer improvements. Not good value for me. YVMV.

The performance improvement in R10.3 would be welcome but I don't tend to hit the CPU limit very often and when I do I just bounce a few tracks down and carry on.

So I'll wait to see if R11 can tempt me. I think it will only be a few months away.

[Heigen5]

By the way, the benchmark file playd the whole file without any glitches in the sound.

[friday]

Yes, yes, yes... so excited about your news. Someone testet with UAD plugins?

[antic604]

Someone testet with UAD plugins?

Aren't they using external DSP from Universal Audio boxes?

[bxbrkrz]

While expecting 110% support for it forever.

I'm a 9.5-er. Not upgrading was purely a cost vs gains decision for me. R10 had hardly anything to interest me. Yet another new synth I don't need, yet another piano I don't need, a few other average instruments. The only items of interest were Synchronous and the smattering of sequencer improvements. Not good value for me. YVMV.

The performance improvement in R10.3 would be welcome but I don't tend to hit the CPU limit very often and when I do I just bounce a few tracks down and carry on.

So I'll wait to see if R11 can tempt me. I think it will only be a few months away.

I understand

[Boombastix]

By the way, the benchmark file playd the whole file without any glitches in the sound.

Did you try to see if hyper-threading on/off made any difference?

[seqoi]

Someone testet with UAD plugins?

Aren't they using external DSP from Universal Audio boxes?

They are but because of how Reason internally worked (64 samples buffer no matter what your ASIO buffer is - it's explained in PH blog) UAD or other DSP plugins had errors and if yo uadded UAD plugin to your Reason session it actually increased your CPU a lot. Which did not make any sense but that's what was happening. This is why person above asked about UAD.

Apparently this is fixed in 10.3 because 10.3 internally is now working differently to old 64 samples topology.

Note i am oversimplifying this and i don't know what PH did for 10.3 - but whatever they did people are noticing improvement - hence my hope UAD plugins is also fixed. And his question isn't error.

[Heigen5]

By the way, the benchmark file playd the whole file without any glitches in the sound.

Did you try to see if hyper-threading on/off made any difference?

I just tested the benchmark file without hyper-threading = Reason playd to the 1:02 and then started to glitch, so yeah, at least in my case the hyper-threading adds at least 26 extra seconds without anything glitching.

[guitfnky]

Did you try to see if hyper-threading on/off made any difference?

I just tested the benchmark file without hyper-threading = Reason playd to the 1:02 and then started to glitch, so yeah, at least in my case the hyper-threading adds at least 26 extra seconds without anything glitching.

interesting, so turning on the hyper-threading option might actually help improve performance now, in 10.3. I don’t know of too many people who get better performance with that option on in 10.2 or earlier.

[Heigen5]

I just tested the benchmark file without hyper-threading = Reason playd to the 1:02 and then started to glitch, so yeah, at least in my case the hyper-threading adds at least 26 extra seconds without anything glitching.

interesting, so turning on the hyper-threading option might actually help improve performance now, in 10.3. I don’t know of too many people who get better performance with that option on in 10.2 or earlier.

Now it seems to help it seems - the Props didn't make that big of noise about 10.3 - at least I am very surprised with the results. I can now delay my plans about to get a faster PC.

[RSWRecording]

Damn. Cool to see the stats!

[esselfortium]

I just tested the benchmark file without hyper-threading = Reason playd to the 1:02 and then started to glitch, so yeah, at least in my case the hyper-threading adds at least 26 extra seconds without anything glitching.

interesting, so turning on the hyper-threading option might actually help improve performance now, in 10.3. I don’t know of too many people who get better performance with that option on in 10.2 or earlier.

It seems to be CPU-dependent. On my i7-4790k performance has usually been better with hyperthreading on.

[EnochLight]

interesting, so turning on the hyper-threading option might actually help improve performance now, in 10.3. I don’t know of too many people who get better performance with that option on in 10.2 or earlier.

It seems to be CPU-dependent. On my i7-4790k performance has usually been better with hyperthreading on.

Right, and with my i7-3770k performance has usually been worse with Hyperthreading on (though in some occasions, it's better). There seems to be no rhyme nor reason to it, so I just try it on and off when my project starts to get heavy.

[Sargie]

I can add to the general chorus of praise. VST Plugins that would barely function before under even a moderate load are now functioning perfectly well with pretty much anything I can throw at them.

Clearly the refactoring of the audio rendering system to increase performance has been a success.

Thanks PH!

[Marco Raaphorst]

Did you try to see if hyper-threading on/off made any difference?

I just tested the benchmark file without hyper-threading = Reason playd to the 1:02 and then started to glitch, so yeah, at least in my case the hyper-threading adds at least 26 extra seconds without anything glitching.

interesting, on my Mac WITH hyperthreading it performs worse.

[Boombastix]

interesting, on my Mac WITH hyperthreading it performs worse.

Make sure you use the SAME TEST file as the OP when you report results. HT is supposed to use unused CPU cycles as a "ghost"/virtual CPU core, but if you have a DSP channel that is very CPU heavy and must be calculated in a single thread then that might be a bottle neck. So, I can see how results seem inconsistent if that is the case. I'm not 100% sure but the HT performance is probably dependent on the actual song/RE/VST setup and if it can be split into more CPU threads or not.

It would be interesting if you all test with HT on/off using the same test file - then we can better get an idea and perhaps even figure out best approach/recommendation for users.

[Heigen5]

I just tested the benchmark file without hyper-threading = Reason playd to the 1:02 and then started to glitch, so yeah, at least in my case the hyper-threading adds at least 26 extra seconds without anything glitching.

interesting, on my Mac WITH hyperthreading it performs worse.

Weird, you have an intel-mac I assume?

[guitfnky]

aren’t all modern Macs Intel-based?

[EnochLight]

Weird, you have an intel-mac I assume?

aren’t all modern Macs Intel-based?

Yeah... there hasn’t been a PowerPC Mac line in over 12 years.

[Heigen5]

Weird, you have an intel-mac I assume?

aren’t all modern Macs Intel-based?

Yeah... there hasn’t been a PowerPC Mac line in over 12 years.

Ok, I'm not an expert about Macs, so thanks for clearing it up. Anyway, anyone wants to guess why Marco's Mac is not taking the advantage of the hyper-treading then?

[antic604]

Ok, I'm not an expert about Macs, so thanks for clearing it up. Anyway, anyone wants to guess why Marco's Mac is not taking the advantage of the hyper-treading then?

There's no such thing as "taking advantage of hyperthreading" for audio processing - it's either the same performance (or perhaps maaaaarginallly better), or much worse:
viewtopic.php?f=4&t=7505905&p=421244

[EnochLight]

There's no such thing as "taking advantage of hyperthreading" for audio processing in Reason - it's either the same performance (or perhaps maaaaarginallly better), or much worse:
viewtopic.php?f=4&t=7505905&p=421244

Fixed that for you.