Single core versus Multicore CPU

[CS70]

2. You also talked about real-time. is it true that real-time processing is done better with more cores (again, assuming no bottlenecks from other PC parts)?

A little point of pedantry - you don't have real "real time processing" in a system unless the operating system (and to a degree the hardware) is "real time" (regular Windows and OSX aren't): true "real time" processing means guaranteed processing time - and since the time-quality-price triangle is valid for CPUs as well , the implication is that in a real time system the same computation is not guaranteed to provide the same result every time. In practice you will get whatever state of the computation exists when the deadline (the time allocated to the operation) expires - and this specific kind of interruption must generally be implemented at very low level, in hardware.

General purpose operating systems prioritize correctness of result over time guarantees, and at most provide an estimate of execution time for a given computational load. Of course we use "real time" loosely to indicate that the system is fast enough in the most cases. In practice, under a load which is easy to handle by the resources available (cores, bandwidth, local memory etc), a non RT OS will in average work fast enough to make little difference, but all the personal computers we use tend to be of that sort.

[wireman]

General purpose operating systems prioritize correctness of result over time guarantees, and at most provide an estimate of execution time for a given computational load. .

Really? I think they prioritise fair access to resources (for some measure of fair) for a running workload.

[CS70]

General purpose operating systems prioritize correctness of result over time guarantees, and at most provide an estimate of execution time for a given computational load. .

Really? I think they prioritise fair access to resources (for some measure of fair) for a running workload.

It's a different level of abstraction. Fair access to resources, for whatever definition of "fair", can be provided by both real time and non real time systems - it's only a matter of load distribution. Both models allow for preemptive scheduling of computational flows.

But the defining characteristic of "correctness"-oriented system is that the flow of control does not return until a computation is completed, regardless how much time it takes. 2+2 is always 4, but it may take a potentially unbounded while for the result to be computed and the flow of control returned to tthe requester.

A "true" realtime system, on the other side, works with strict deadlines: if the allocated time for the computation is expired the flow of control and the result is returned (possibly with a flag set). So 2+2 occasionally maybe 3.5 or whatever.

This may seem odd but typically in physical control systems is much better to have an occasionally slightly wrong result than a possibly long waiting time before the computation is finished. This is also occasionally useful for algorithms which return quickly most of the time but may behave different for a few sets of input parameters, etc.

Obviously a real time system must be sized so that the majority of the expected computations do not generate errors.. and that's why it's not so suited for general purpose computers. Both models have equivalent computational power and you can emulate one with the other but when implemented in a physical machine, you've gotta choose.

[wireman]

General purpose operating systems prioritize correctness of result over time guarantees, and at most provide an estimate of execution time for a given computational load. .

Really? I think they prioritise fair access to resources (for some measure of fair) for a running workload.

It's a different level of abstraction. Fair access to resources, for whatever definition of "fair", can be provided by both real time and non real time systems - it's only a matter of load distribution. Both models allow for preemptive scheduling of computational flows.

But the defining characteristic of "correctness"-oriented system is that the flow of control does not return until a computation is completed, regardless how much time it takes. 2+2 is always 4, but it may take a potentially unbounded while for the result to be computed and the flow of control returned to tthe requester.

A "true" realtime system, on the other side, works with strict deadlines: if the allocated time for the computation is expired the flow of control and the result is returned (possibly with a flag set). So 2+2 occasionally maybe 3.5 or whatever.

OK, yes. For some reason when I read the post I thought you meant that a general purpose OS could choose how much priority to place on correctness.

[CS70]

OK, yes. For some reason when I read the post I thought you meant that a general purpose OS could choose how much priority to place on correctness.

All good, no stress.

[LoveMonkey3833]

Turbo boost requires sufficient power and cooling available to maintain the boost. If the system has this is can be maintained indefinitely.

Would a stock cooler do that?

What chip would you go for? Out of the high end choices on this thread I'd go for a Ryzen 3900x because it has speed and cores.

@JacoVanDujin Put simply clock speed gives low latency and more plugins on one track. More cores gives more tracks or virtual instruments.

DO NOT USE A STOCK COOLER!!!!!
Get a Noctua DH-15 cooler with Kingpin cooling paste! Mixing and doing audio work makes your computer hot as it maxes everything out! Stick with air cooling as it's easier and tbh most water coolers that you buy off the shelf still don't stand up to Noctua's top of the line air cooler the NH-D15. It's the best air cooler in the biz. There are 2 that come close right now, maybe 3, but still can't beat it. Also get a case that has great air cooling, but isn't crazy loud! Like Rosewill Thor V2! Nothing beat's it that I've come across yet!
Make sure you've got a great PSU as well, like an Evga Supernova G2 1300. People will tell you to get maybe a 650 PSU or some bs like that, but don't take into account all the crap we have to install in the computer, plus all the peripherals we run off of USB hubs, plus the addon's we might put on later! Get an overpowered PSU, and make sure it's at least Gold standard, and it has a 10year warranty. You don't want your thousands of dollars setup to take a dump on you and blow up your shit! Trust me it's not worth it. Had a Rosewill PSU that literally blew up my AMD build back in say 2010 or so. Their PSU's suck, but cases are great! Also, get extra fans for your case, yes this case is the only one that comes with a huge ass 280mm fan(or something like that)on the side blowing directly onto your CPU, but you still want extra fans in there! I'd get Noctua fan's or Arctic fan's or whoever bought out Scythe....them guys. Really I'd skip Artic and just go Noctua or the Scythe wannabe guys. Cooling is super important! VERY VERY VERY Important! If that didn't sink in, let me say it again in caps....COOLING IS VERY IMPORTANT!
Oh and the AMD 8650 (or whatever number it is)on all the tests smoked the Intel CPU's. But In my experience AMD CPU's suck ass when it comes to audio processing. Maybe the new ones are killer, but until I got concrete evidence I stuck with Intel. Recently upgraded my cpu from a 6 core 5820k to a 9700k. I wanted to go with the I9 but couldn't afford it since I'm building from scratch. Didn't have the extra $227$ to throw on a cpu with marginal improvements. Of course, they were out of the AMD ones the 50's whatever number they are. But I wanted Thunderbolt compatibility and the Gigabyte Z390 Designare board was the only one proven to be compatible with UAD's Thunderbolt. AMD didn't have any boards that were "proven" to be working on Thunderbolt without issue on a Windows-based machine. Now that may have changed in the last 2 weeks since I did my research, or I may have missed it, but still, Intel has traditionally been better for audio production. But I totally hate that since I think they are too damn expensive for what you get and would rather have 12 cores that are faster with AMD....and generally cheaper too. Alas, it wasn't meant to be, and I probably should be grateful because my experience with AMD as audio production workstations is absolutely terrible. Nothing but issues at every turn! Although I would recommend an AMD GPU instead of Nvidea for your audio production CPU as there are a few articles of how AMD beats Nvidea somehow for audio production computers. You'll have to google it as I'm too lazy right now to find it!
My recent build is an Intel 9700k I7, Noctua NH-D15 Chromax cooler, Corsair Vengence LPX 3200 ram, AMD Sapphire 590x GPU(the blue one) got it open box for $160 at Microcenter, an Adata-3000 write speed or so nvme SSD drive(don't remember the numbers was $100 on sale though), Gigabyte Z390 Designare mobo, Evga Supernova G2 1300 PSU, Rosewill Thor V2 case, and about 7 additional Noctua fans in there. All of it was overclocked to 4.9 and the ram was as well, but a buddy did that for me and made it so that I could do change from normal function to overclocked audio production function from my desktop after a restart. So if I'm not working on a crazy large project I don't have to worry about it getting crazy hot in there, although my cooling is pretty wicked good for not going the water route. Oh and all my fans including my cpu cooling fan is hooked up to 2 rotary dials on the front of my case so that while recording I can turn down the fans, and then crank them when mixing! They can get loud when cranked even though I got quiet fans, nothing you can do about moving that much air. Think of jet's pushing tonnes of air....Not the same but you can't make it dead quiet either no matter what you do. Only mitigate it. Buy buying a case like a Be Quite case with soundproofing in the case, you'd have to go with water cooling as not enough air would be able to get in the case otherwise! Good luck on your build!

[Pete Kaine]

A "true" realtime system, on the other side, works with strict deadlines: if the allocated time for the computation is expired the flow of control and the result is returned (possibly with a flag set). So 2+2 occasionally maybe 3.5 or whatever.

Aye, it's for want of a better term really. Given ASIO is buffer based, it's not quite real time of course, although in contrast expired time in this context results in data errors manifesting as currupted audio whilst still continuing to otherwise process as normal.

It's a far cry from a off-line job through as the demands remain different, we need an intermedite term for it.

[James Perrett]

It's a far cry from a off-line job through as the demands remain different, we need an intermedite term for it.

Back in the days when I was dealing with real time operating systems (like VxWorks or VRTX) we would talk about hard real time systems and soft real time systems. The word 'deterministic' seemed to crop up quite a bit too.

[CS70]

A "true" realtime system, on the other side, works with strict deadlines: if the allocated time for the computation is expired the flow of control and the result is returned (possibly with a flag set). So 2+2 occasionally maybe 3.5 or whatever.

Aye, it's for want of a better term really. Given ASIO is buffer based, it's not quite real time of course, although in contrast expired time in this context results in data errors manifesting as currupted audio whilst still continuing to otherwise process as normal.

It's a far cry from a off-line job through as the demands remain different, we need an intermedite term for it.

No worries, I was just being pedantic because it was a good opportunity to shed some light on a computer science idea which is not widely understood. I use myself "real time" in the same context, because as you say the context makes clear what one means..

[aarav]

Hi.
I would've posted this on a separate thread but can't seem to find out how

Which one of these would perform better generally in fl studio?
i7 10700k 8 cores - Scored 12% better in multi core
vs.
Ryzen 5 5600x 6 cores - Scored 12% better in single core performance

Better multi core performance with two extra cores vs. Better Single core performance with two cores less

I would appreciate any input.

[Yearofthegoat]

Not much between them, but at the moment Scott over on the FL Studio forums seems to be going with Ryzen:
https://forum.image-line.com/viewtopic. ... x#p1581110

And as has been mentioned on this thread, all things being even, go for the higher single-thread score.

My Ryzen 5 3600 clock is highly dependant on cooling. When testing fans I could literally see the clock speed climbing as I increased the CPU fan speed.