Awesome. Could you state what kind of mhz drop off you get when using the 2 kits? Will a 5200mhz drop to say... 2000? Or just 4500? If it's the latter i'm not too worried!
Solid reporting, my friend! Exactlywhat im dealing with now too. And there's next to no muso / studio tubers talking about this stuff. Keep up the good work.
You're doing a great job on reporting DAW performance man! I haven't seen anyone else do this. Please keep this up! My question is: Does more CPU cache help? Ryzen 5800X vs 5800X3D for instance. Also looking forward to future Zen4 V-Cache vs Intel 13th gen benchmarks if you end up doing them. Cheers.
I have checked with Puget Systems and they have a i7-13700k with ASUS Z-690 ( I intend the pro art creator even though Z790 is in the mist somewhere)) running with DDR5 ram claiming 4800GHz. they don't say 'what' ram they are using but they disarm the AI and claim 4800 is possible. Do they set the 4800 as a target and a brick wall or are the DDR5 chips actually going to be fully efficient at these speeds with 64GBram?? . I am doing Home recording studio and I'd like 64 gig( in the 2 stick config) DDR5 ram. Can you make any suggestions as to what currently, (November 11, 2022), is the most advisable DDR5 ram running virtual instruments ?? I want to minimize all the 'overclocking as much as possible but still want to get the value out of the DDR5
Love your videos guys! Some great future topics might be the noise floor or levels of EMI/RFI of various motherboards. I imagine almost noone is interested in/can afford the new w2400 HEDT Xeons? I was eagerly awaiting them after following the leaks and rumours for the past couple of years - until the prices were revealed!
I am in the process of specing out a 13700k build for music production (cubase), and was wondering what kind of a performance difference will there be between ddr4-3200 to ddr5-5600? (2x16 if that matters)
Just wondering, really what does the timing and mhz improve on in DAW use? I'll be upgrading soon from 1600mhz and a 2nd generation i7... so anything is going to be a huge improvement. But i feel like i'm worrying too much about the mhz when its capacity that matters far more when using orchestral libraries. Who cares about speed if you simply dont have enough ram. And what does the speed do in real time performance? Load libraries a few ms quicker?
Memory speed has a direct correlation to polyphony. The faster the memory, the more notes you can play back simultaneously from a plugin like kontakt. But cpu still plays a huge part of that too. However, the jump you’ll have from a i7 2700K to a 12900K is going to be the in the ballpark of a difference of playing back 6 or 7 times the amount you can do now. Don’t worry about it, grab whatever is a good deal and get on with writing some music without having to bounce to audio as much.
@@ScanProAudioTV oh right thanks for the info on ram speed. I dont seem to have a problem with polyphony on 1600mhz at the moment to be honest I just max out the GB easily before I hit that limit as I only have 16gb!
@@MichaelDowComposer Be aware that the Vi-test is a very extreme case of only kontakt instances where an extreme amount of samples are streamed in a way that never ever will be the case for anybody creating a song. Yes if you are into the absolute biggest imaginable libraries only using Kontakt and maybe working for movie productions then you may possibly gain just a little from better memory, but even then it would be nowhere this "synthetic" Vi-test. But for a more normal "mixed usage" where you also lean more heavily to "pure" vst-instruments that lean extremely heavy on the CPU for rendering (not pushing around samples between memory banks), then you are much, much better off looking at what the DPS-bench parts show. And there the difference between DDR4 and DDR5 is zero! It doesn't exist at all. I think this is the weakest point of DAWbench in the way it names the Kontakt-test "Vi" as if it is only a test of virtual instruments in contrast to "DSP" as if that is totally opposite of "virtual instrument " performance. One use-case for mixing or mastering ("DSP") and another for "Virtual instruments" (feels more like a home user). By that, it may also be misleading for users mainly interested in normal synths as vst:s and therefor buy only Intel (Which always does much better than AMD in the "Vi-test"), when they often could have spared money and in many cases had a better or at least equal performance for more purer DSP-cases as a pure virtual instrument without streaming of large databanks of samples to and from memory is! Money could also be spared in just buying stock bog standard DDR4 RAM at the cheapest prices since in almost all benchmarks for almost everything else that is involving real-time performance, it shows a very little difference, if any at all! And my hunch is that if you do a mixed test with normal song writing, with different types of virtual Synths (not RAm dependant), some Kontakt instances, some effects, then you will see very little benefits from different Ram type or even different Ram speeds.
How fast is fast these day’s?, I’d love to see a bench test now compared with yesterday. 😊 what’s a 10 year gap given us. DDR4/5, SSD’s NvME, multithreading performance on processors. Against HD, DDR3/DDR4 and older processors. What’s the relative uplift I wonder?
It's quite easy to check CPU vs CPU. But I have no idea if what you're asking can be tested on some comparison site, perhaps one exists though. Would be really interesting to see results. I have been told the 6700k in my laptop to 12900k is - multicore around 4.5X and the single core score is around 80% higher.
Does 32gb sticks vs 16gb sticks matter when working with virtual instruments? Is it better to have 16gb x 4 ,or will 32gb x 2 have same performance? I will need 32gb x 4 in the end on a micro atx motherboard for ryzen 7000
Hey, I just saw your sm7b and mv7 (with new pop filter) review, can you test out the new beacon mic and tell me (not minding about price) if the mv7 is still a better choice?
Well, I think I will just wait till these Z690's can handle 64-128GB at better speeds. Not that I've ever had much success with my memory running at supposedly plug-n-play OC speeds glitch free, I've always had to set it back to slower speeds, so maybe I'm just tilting at windmills here.
I am looking to build a new machine and seems a little early to be buying ddr5 motherboard and ram. Would it be more sensible to get a ddr4 ram and motherboard at the moment? Thanks
Curious if and when the dawbench results for the new m1 studio max and ultra will be reviewed. Currently wondering if it is better to buy this than have to worry about the issues with hackintoshes and alder lake, big.little, ddr5 and no native support maybe ever now with the new m1's. Whats the point of getting the 5950x and 12900k if half the cores won't work. i know alot of guys use reaper and windows but there is tons of us guys that use Logic Pro X so using osx on a hackintosh is necessary. Thanks for these dawbench comparisons and for all the work you do!
Its a bit hard as there is a cross platform difference to the plugins, so its all a bit vague, but we certainly know ballparks. M1 Max is somewhere roughly around the i5 12600 level and the Ultra is performing above the 12900K, 5950X and the X299 chips (so above everything then...) in preliminary DSP tests. This will change when the new zen 3 threadripper chips and AMD 7000 series get released.
Update : 2x32GB - 64GB - 5200MHZ packs are now available.
Awesome. Could you state what kind of mhz drop off you get when using the 2 kits? Will a 5200mhz drop to say... 2000? Or just 4500? If it's the latter i'm not too worried!
Solid reporting, my friend! Exactlywhat im dealing with now too. And there's next to no muso / studio tubers talking about this stuff. Keep up the good work.
You're doing a great job on reporting DAW performance man! I haven't seen anyone else do this. Please keep this up!
My question is: Does more CPU cache help? Ryzen 5800X vs 5800X3D for instance.
Also looking forward to future Zen4 V-Cache vs Intel 13th gen benchmarks if you end up doing them.
Cheers.
How influential is the 13900K v the 12900K in the virtual instrument test?
Can't wait to see your Raptor vs Zen4 testing charts, sir!
thank you guys !!!!
Thanks for the video! Would love to see a follow up video about Ram and what to get now! Cheers
I have checked with Puget Systems and they have a i7-13700k with ASUS Z-690 ( I intend the pro art creator even though Z790 is in the mist somewhere)) running with DDR5 ram claiming 4800GHz. they don't say 'what' ram they are using but they disarm the AI and claim 4800 is possible. Do they set the 4800 as a target and a brick wall or are the DDR5 chips actually going to be fully efficient at these speeds with 64GBram?? . I am doing Home recording studio and I'd like 64 gig( in the 2 stick config) DDR5 ram. Can you make any suggestions as to what currently, (November 11, 2022), is the most advisable DDR5 ram running virtual instruments ?? I want to minimize all the 'overclocking as much as possible but still want to get the value out of the DDR5
Interesting! My order for the new 'Dr Mix' 12900K has been in for just a few days - chose the DDR5 5600 after seeing the last video. 😎 🤓
hello! can we have the difference between 1 stick (dual channel) vs 2 sticks (quad channel) vs 4 stick if possible (quad channel) please?
He's gonna need a Threadripper platform to test that. But I would really like to know that as well.
Thanks so much for all the info! Very, very helpful and appreciated! Have you by chance tested 6000mhz ram yet?
Love your videos guys! Some great future topics might be the noise floor or levels of EMI/RFI of various motherboards.
I imagine almost noone is interested in/can afford the new w2400 HEDT Xeons? I was eagerly awaiting them after following the leaks and rumours for the past couple of years - until the prices were revealed!
four sticks of ddr5 gonna give the speed promsed now ?
Just wondering, is that problem now resolved?
I am in the process of specing out a 13700k build for music production (cubase), and was wondering what kind of a performance difference will there be between ddr4-3200 to ddr5-5600? (2x16 if that matters)
Intel Gen12 and 13 CPUs have performance issues with cubase.
Just wondering, really what does the timing and mhz improve on in DAW use? I'll be upgrading soon from 1600mhz and a 2nd generation i7... so anything is going to be a huge improvement. But i feel like i'm worrying too much about the mhz when its capacity that matters far more when using orchestral libraries. Who cares about speed if you simply dont have enough ram. And what does the speed do in real time performance? Load libraries a few ms quicker?
Memory speed has a direct correlation to polyphony. The faster the memory, the more notes you can play back simultaneously from a plugin like kontakt. But cpu still plays a huge part of that too.
However, the jump you’ll have from a i7 2700K to a 12900K is going to be the in the ballpark of a difference of playing back 6 or 7 times the amount you can do now.
Don’t worry about it, grab whatever is a good deal and get on with writing some music without having to bounce to audio as much.
@@ScanProAudioTV oh right thanks for the info on ram speed. I dont seem to have a problem with polyphony on 1600mhz at the moment to be honest I just max out the GB easily before I hit that limit as I only have 16gb!
@@MichaelDowComposer Be aware that the Vi-test is a very extreme case of only kontakt instances where an extreme amount of samples are streamed in a way that never ever will be the case for anybody creating a song.
Yes if you are into the absolute biggest imaginable libraries only using Kontakt and maybe working for movie productions then you may possibly gain just a little from better memory, but even then it would be nowhere this "synthetic" Vi-test.
But for a more normal "mixed usage" where you also lean more heavily to "pure" vst-instruments that lean extremely heavy on the CPU for rendering (not pushing around samples between memory banks), then you are much, much better off looking at what the DPS-bench parts show. And there the difference between DDR4 and DDR5 is zero! It doesn't exist at all.
I think this is the weakest point of DAWbench in the way it names the Kontakt-test "Vi" as if it is only a test of virtual instruments in contrast to "DSP" as if that is totally opposite of "virtual instrument " performance.
One use-case for mixing or mastering ("DSP") and another for "Virtual instruments" (feels more like a home user).
By that, it may also be misleading for users mainly interested in normal synths as vst:s and therefor buy only Intel (Which always does much better than AMD in the "Vi-test"), when they often could have spared money and in many cases had a better or at least equal performance for more purer DSP-cases as a pure virtual instrument without streaming of large databanks of samples to and from memory is!
Money could also be spared in just buying stock bog standard DDR4 RAM at the cheapest prices since in almost all benchmarks for almost everything else that is involving real-time performance, it shows a very little difference, if any at all!
And my hunch is that if you do a mixed test with normal song writing, with different types of virtual Synths (not RAm dependant), some Kontakt instances, some effects, then you will see very little benefits from different Ram type or even different Ram speeds.
How fast is fast these day’s?, I’d love to see a bench test now compared with yesterday. 😊 what’s a 10 year gap given us. DDR4/5, SSD’s NvME, multithreading performance on processors. Against HD, DDR3/DDR4 and older processors. What’s the relative uplift I wonder?
It's quite easy to check CPU vs CPU. But I have no idea if what you're asking can be tested on some comparison site, perhaps one exists though. Would be really interesting to see results. I have been told the 6700k in my laptop to 12900k is - multicore around 4.5X and the single core score is around 80% higher.
is it possible to do a benchmark on higher DDR5 speeds? 6000MHz 6400MHz 6800MHz 7000MHz
so after 1 year is ddr5 able to get the speed it promises ?
Thanks for the useful comparison. What speed was the DDR3 tested?
Does 32gb sticks vs 16gb sticks matter when working with virtual instruments? Is it better to have 16gb x 4 ,or will 32gb x 2 have same performance? I will need 32gb x 4 in the end on a micro atx motherboard for ryzen 7000
Hey, I just saw your sm7b and mv7 (with new pop filter) review, can you test out the new beacon mic and tell me (not minding about price) if the mv7 is still a better choice?
Well, I think I will just wait till these Z690's can handle 64-128GB at better speeds. Not that I've ever had much success with my memory running at supposedly plug-n-play OC speeds glitch free, I've always had to set it back to slower speeds, so maybe I'm just tilting at windmills here.
When will you bring the 2023 PC builds, with the latest i9s and 7950x? 😬
Tests have been done, should be a vid in a couple of weeks.
I am looking to build a new machine and seems a little early to be buying ddr5 motherboard and ram. Would it be more sensible to get a ddr4 ram and motherboard at the moment? Thanks
Any updates on this interesting topic? Have you encountered any issues running 2 sticks of 32gb DDR5 at full capacity?
Certainly the Corsair 2x32GB packs are working fine. Try getting 4 32GB sticks working stably still isn’t, even considerably clocked down.
@@ScanProAudioTV Great to know. Thanks for the feedback :)
Curious if and when the dawbench results for the new m1 studio max and ultra will be reviewed. Currently wondering if it is better to buy this than have to worry about the issues with hackintoshes and alder lake, big.little, ddr5 and no native support maybe ever now with the new m1's. Whats the point of getting the 5950x and 12900k if half the cores won't work. i know alot of guys use reaper and windows but there is tons of us guys that use Logic Pro X so using osx on a hackintosh is necessary. Thanks for these dawbench comparisons and for all the work you do!
Its a bit hard as there is a cross platform difference to the plugins, so its all a bit vague, but we certainly know ballparks.
M1 Max is somewhere roughly around the i5 12600 level and the Ultra is performing above the 12900K, 5950X and the X299 chips (so above everything then...) in preliminary DSP tests.
This will change when the new zen 3 threadripper chips and AMD 7000 series get released.
You guys ever going to fix your website so that I can access it from North America without needing to use a VPN?
Hello there, which z690 model would be the best for guitar recording and production ?
z690s are better for drums.
b660s are more suited to record a guitar.
The one with adequate vrm to power your processor. There's videos on it. Msi is good this generation in general. ASRock is typically not so good.
Bollocks . DDR5 and z690 running good nothing wrong with it