How far have computers come in the last 5 years? For DAW usage?

Audio Plugin Hosts and other audio software applications discussion
RELATED
PRODUCTS

Post

Nah man.
2500k and youre all set.
ImageImageImageImage

Post

exmatproton wrote:
acYm wrote:my 2500k will be what, 6 years old soon, I'm sloooowly starting to feel like upgrading it. there's no speed increase with newer ones, which is really lame, but the larger caches and ddr4 bandwidth are interesting.
huh??

I run a 4970k and this one has an increased speed compared to yours.

http://cpuboss.com/cpus/Intel-Core-i7-4 ... e-i5-2500K

Also, double the threads will (especially with multicore support in DAW's) increase your performance even more!
in the realm of overclocking and single-core usage, which is what I care about, no current cpu justifies the upgrade cost. I reach 4.6 with my chip. I'll be more interested when something hits 6.0, or 5.5 at least.

Post

I recently wrote this for the FL Studio manual...and thought it may be relevant to interests in this thread. The issue boils down to single-core performance. This is absolutely key to all DAW software. Unfortunately single core processing speed has taken a back seat to multi-core performance in the last 5 years (to a certain extent at least).

If you open the system CPU meter and examine the number of cores used and their relative loading you may wonder why your CPU seems to be under-utilized. This may come as a surprise since some marketing departments make a lot of noise about their multi-core 'optimizations'. What controls core assignment? - The OS system 'Scheduler' is responsible for the core assignment, not the DAW. To complicate matters, logic dictates that many tasks can't be processed in parallel. Let's look at this in some more detail by considering Metrics, time and logic:

Different metrics - Most DAW CPU meters measure how quickly each audio-buffer is filled. CPU = (Time taken to fill the buffer / Buffer length). The Operating System measures overall CPU and core 'utilization'. Utilization shows the portion of processing-slots on the CPU that the CPU scheduler assigned to all programs and the OS. Utilization = (processing-slots filled / processing-slots available). As the CPU has a finite number of 'slots' where code/threads can be processed, utilization is more about how close to maximum capacity the CPU was, not how quickly it was processing any given task (such as the audio mixing thread). Important to remember, we refer to this idea below. The two measures are related, since a CPU that can process a lot of tasks at once is probably able to fill the audio-buffer more quickly than one that can't. However, there's a lot more to audio processing than just CPU capacity to do work, and the two measures are definitely not the same thing...even if they are both reported in percent! CPU utilization is not a measure of how quickly or when a specific task was processed, and that leads us to consider the ideas of time and logic.

Time-scales - DAW CPU meter is relative to the audio buffer size, 1 ~ 50 ms. The OS CPU utilization meter works on a 1000 ms interval. The difference in time-scales is at least a 20x and up to 1000x. This is very important to remember. While the operating systems CPU meter may show 30% utilization, over the last 1000 ms, there may have been multiple occasions during that period where real-time audio processing experienced interruptions. Why? If real-time audio 'Mixer threads' (packages of work for the CPU), have to wait on other threads to finish, because they can't be multi-threaded (processed at the same time), the DAW may experience audio underruns, or at least very high DAW CPU meter readings. At the same time, the OS may report low overall and or individual CPU utilization. The CPU could have done a lot more work than it did, if it had something else to do at the same time. The reality for audio processing is the CPU must often wait for program and system related tasks to complete before it can continue, and so, may struggle to keep up with the very high demands of real-time audio output (generating 44100 samples per second, on an ongoing basis, without an interruption of a single sample, 0.02 ms). Just why the CPU must 'wait' is all to do with logic:

The logic of audio processing - There are long lists of tasks that must be processed in sequence, and this means logically can't be simultaneously multithreaded. For example: Plugins must wait for instructions from the Piano roll and Playlist before they make sound. Effects must wait for the audio stream from upstream instrument plugins before they can process it. Further, it's not possible to parallel-process (multi threaded) instruments and FX that are on the same Mixer channel (their audio is mixed together), or even in the same Mixer routing pipe-line (when one Mixer track is linked to another and another, even FX processing has an order from top to bottom in the FX stack). Then, the Master Mixer track must wait for every instrument > mixer track > effect to be processed before it can process the audio through the Master effects. So logically, there is a lot of waiting that is a natural and unavoidable fact of DAW music processing. Think of a production line with each station waiting for the one before it. This means the CPU may not be particularly busy, using all its cores and processing slots, yet it runs out of time to fill that tiny 5 ms audio-buffer because there was a lot of waiting for things that needed to be processed in sequence. It should be clear that fast processing is very important and this is not the same thing as multi-core processing. TIP: When comparing CPUs, look for the fastest single-core performance scores in a package with at least 4 physical cores. Most CPU benchmarks list single core performance. For example, CPU mark lists the single core score below the multi-core score.
Image-Line are proud developers of - FL Studio, FL Studio Mobile & Audio Plugins.

Post

Yep. No, I don't mean I'm as smart as "image-line" what I mean is that benchmarks ARE NOT real-world performance.

Ok, my 4770k is better than my orginal 860 (first gen) but it's not the world-beater you'd think.

Now, more to this point, every single host is VASTLY different than the other. To my surprise, reaper is still king. Some are fine (live, sonar once you get past the first core they generally taper off) and then reason is "ok" and some others do what they do......but no two hosts are the same.

OPEN LETTER:

Please all hosts, optimize and make this NOT A THING anymore. ;)

BTW, after YEARS of owning and trying FL (with this bizarre thing of the faster my computer got, the more FL would increase IT'S cpu :lol: ) FL 12 with minor and easy tweaks, with WINDOWS 10 actually worked fine. Just wanted to say that for the record.

Post

Mine came about 30 kilometres..but not as the crow flys (flies?), it was in the back of my car and I took the twisty way.
Last edited by werp on Tue Jul 12, 2016 6:44 am, edited 1 time in total.
I'm tired of being insane. I'm going outsane for some fresh air.

Post

The external sound cards that we use, they should start coming with APUs. The same way graphics require GPUs.
"The educated person is one who knows how to find out what he does not know" - George Simmel
“It's what you learn after you know it all that counts.” - John Wooden

Post

Lets say we get to the point where we have like 50+ cores on a chip.
Or at least, lets say "progress" continues in that direction, and we are left with minimal core speed increases, but massive core number increases;
Is it feasible that the way audio is processed could be re-imagined to use more cores? Like, massive latency, and something like a whole core working as a 'traffic controller' for the mis-aligned streams going through the other cores. One core, acting as a two-bus, and reporting latency.
Obviously, I know nothing. :)
ImageImageImageImage

Post

To summarize: software has evolved enough to (mostly) legitimate an update of your favorite worksstation software.
Hardwarewise however it depends. Progress had sort of stalted. Moore's law has come to an end. Microprocessors five years ago were only marginally slower than today's top models. I hate it when I see a new processor generation having only an extra 10% of computing power.

My personal law to decide if a hardware upgrade is desirable is: does it give me at least a doubling of performance? If not I will stay with my old setup which will serve me enough to wait for real improvements hardwarewise. Waiting will itself pay off. One day microprocessors will be fast enough to legitimate an update. Hopefully AMD's upcomming ZEN-processor will take the lead.

That said, if I will never upgrade my DAW anymore (both hardware and software wise) I will still be in a position to create great and innovative music anyhow. Creativity does not (totally) depend on the speed of your machine altough a speedy machine can be a great joy. :-)
So, while waiting for technology to push the limits of computing further and further, just make music as if there will never be any upgrade anymore.
The more I hang around at KVR the less music I make.

Post

10 core i7 CPUS are now available at a huge cost though. Other than that I'm not tempted to upgrade my first gen i7. I think I would get more CPU spare if I spent the same money on a Dave Smith OB6 desktop replacing one of the DIVA tracks.

Post

There is never an end. Never. I don't care how f**king powerful this shit gets, no end.

Post

incubus wrote:There is never an end. Never. I don't care how f**king powerful this shit gets, no end.
Upgrade or die bitches ! :lol: (My i7 5930 already about 2 years old, it's sucking ass in some ways).

Post

Codestation wrote:
incubus wrote:There is never an end. Never. I don't care how f**king powerful this shit gets, no end.
Upgrade or die bitches ! :lol: (My i7 5930 already about 2 years old, it's sucking ass in some ways).
Mine's older (4710HQ), and quite an improvement over my 2011 i5...though i'm sure i could've easily buried it the day i brought it home. It's still working for me for the time being (mostly because, well, 3 hard drives in a laptop!). 17" HD display, 3 1TB HD's and 32GB RAM for $2k-ish...just for the expense, it'll have to last a while.

At the time, I was considering going for a dualie xeon (16 core) desktop for another grand...but...just love my laptops.
Feed the children! Preferably to starving wild animals.
--
Pooter | Software | Akai MPK-61 | Line 6 Helix | Dynaudio BM5A mk II

Post

Well, like I said, a) there is NEVER an end! And b) (from earlier) from my first gen i7 to this (4770k) it wasn't the upgrade I was hoping for. At the time the 4770k was pretty much the de facto chip. It was considered fantastic w/o the crazy priced version (3970????? Can't remember)

Anyways, is what it is, certainly doesn't suck, but unless you drop mad coin, seems that we are kinda in a holding pattern.

Post

Hardware has evolved, but not in a direction that would benefit, us music producers the most and at the same time software stagnated, and when I say that I mean software is not keeping up with hardware. Most software is not well optimized for multi core CPUs. Also most software still work on old machines. At the moment software isn't pushing hardware anymore. I the past we needed more CPU power because there was new software that asked for more, right now that's not the case anymore. We have power, but we don't have software to be able to use all that power.

Post

There are long lists of tasks that must be processed in sequence, and this means logically can't be simultaneously multithreaded. For example: Plugins must wait for instructions from the Piano roll and Playlist before they make sound. Effects must wait for the audio stream from upstream instrument plugins before they can process it. Further, it's not possible to parallel-process (multi threaded) instruments and FX that are on the same Mixer channel (their audio is mixed together), or even in the same Mixer routing pipe-line (when one Mixer track is linked to another and another, even FX processing has an order from top to bottom in the FX stack). Then, the Master Mixer track must wait for every instrument > mixer track > effect to be processed before it can process the audio through the Master effects. So logically, there is a lot of waiting that is a natural and unavoidable fact of DAW music processing. Think of a production line with each station waiting for the one before it
Well yes, but why that's an issue? the only reason why we would like to speed up processing is when there are already many tracks with computationally expensive plugins -in particular synths. Different tracks can be easily processed in parallel, while suming their outputs in a bus or master channel is trivial.

In real-world situation you will have many more independent tracks that processor cores :wink: While if you have only few, it doesn't matter and will run fast anyway.
Blog ------------- YouTube channel
Tricky-Loops wrote: (...)someone like Armin van Buuren who claims to make a track in half an hour and all his songs sound somewhat boring(...)

Post Reply

Return to “Hosts & Applications (Sequencers, DAWs, Audio Editors, etc.)”