Memory speed matter in Reason ?

[Voyager]

Hey guys,

So after my thread here viewtopic.php?f=4&t=7506283&p=385248#p385248 i finally went with a i7-4930K. Now my motherboard can handle quad-channel memory and i'm going to get 4x4gb for a total of 16gb.

Now i can get the memory in 1866mhz or 2133mhz for just 10€ more..and both are in CL9.

Intel suggest that the 4930K support up to 1866mhz. So should i go for the sure path and stability with the 1866mhz or i should go with the 2133mhz and hoping it's going to be a valuable ressource in Reason ?

So how do Reason is affected by memory speed and do it matter or not ?

[Oquasec]

Had 2.1ghz ram for about half a decade now, might make difference. I know the amount does tho.

[Voyager]

Anyone else have an opinion on the subject ?

[EnochLight]

So how do Reason is affected by memory speed and do it matter or not ?

Anyone else have an opinion on the subject ?

It's negligible. Reason is mostly affected by CPU speed. That difference in RAM will be hardly noticeable in Reason, if at all. CPU speed and SSD speed have the biggest effect overall.

That said, if the cost isn't that much more for the faster RAM, it can't hurt. It's always nice to squeeze some more points out of a PassMark benchmark or FPS's in a game.

[demt]

timer resolution app https://timerresolution.soft112.com/download.html

imerResolution is an application to change the resolution of the default windows timer. The standard timer on Windows XP can vary between 10 and 25 milliseconds. Therefore if your code uses a timer or sleep value less than the timer resolution on your system you won't be getting the results you expect.
The timer resolution will be globally changed while the application is running and will revert to the previous value when it is closed.
1ms timers are achievable using this application

[demt]

ahhh works on windows 10

[EnochLight]

timer resolution app https://timerresolution.soft112.com/download.html

imerResolution is an application to change the resolution of the default windows timer. The standard timer on Windows XP can vary between 10 and 25 milliseconds. Therefore if your code uses a timer or sleep value less than the timer resolution on your system you won't be getting the results you expect.
The timer resolution will be globally changed while the application is running and will revert to the previous value when it is closed.
1ms timers are achievable using this application

Can you share what this has to do with the OP's post? I haven't heard about this being used for many years...

[4filegate]

Memory, looks like this, for example: "7-7-7-7-7-18". This information refers to the reaction time of the RAM module. The lower the values, "the fever".

[EnochLight]

Memory, looks like this, for example: "7-7-7-7-7-18". This information refers to the reaction time of the RAM module. The lower the values, "the fever".

I'm familiar with what CAS latency/timings is; just didn't realize that was the purpose of that app. I haven't heard it mentioned at overclocking forums I frequent in like 7 or 8 years.

[demt]

still works!

[4filegate]

We don't know what's more important, to have more (GB) Ram or to have faster (MHz + CL-Timing) RAM ... only the programmers of Reason should be able to tell us what's better fast or just more RAM

[EnochLight]

We can say with certainty that the CAS speed of RAM has little to no effect with Reason. It's negligible. We can make this assumption just by comparing CAS timings of various RAM in gaming performance, which is where it's really noticed (fun fact: it's often negligible in gaming). There's a lot of data available in gaming and CAS timing, and usually we're talking a difference of just a few FPS's. The clock rate, on the other hand, will affect the overall speed of your OS/system in general, so faster clock rate is likely more important.

Reason is not - nor has it ever been - a RAM-intensive application. It's always been CPU-intensive, and to a lesser degree, hard drive-dependant. Even in the popular benchmarking threads here in this forum, we've seen that adding more RAM has a very negligible effect on Reason's performance (say for instance, stepping up from 8 GB to 16 GB). That said, if you tend to use a lot of samplers (which keep their data stored in RAM), the more RAM the better!

[Voyager]

We don't know what's more important, to have more (GB) Ram or to have faster (MHz + CL-Timing) RAM ... only the programmers of Reason should be able to tell us what's better fast or just more RAM

I couldn't work with anything else except Reason and as far i love it i don't like Propellerhead lack of presence and ignorance in this forum to clear up some things about some unanswered technical topics and miscellaneous customers questions.

And what make it worse is that they are lurking this forum every single day...

They can feel very, very lucky to have some nice and kind people here that do a part of their job for free and spend a crazy amount of time by answering and helping people out there with their questions and problems. Hats off to those guys !

[4filegate]

I don't want to be shown a race interrupt on the DAW Reason by the propellerheads programmers, Eh!

Our empirical test can provide an answer to this question! We are carried out an empiric investigation (soon)

[normen]

Memory speed is very much important for a DAWs speed, Enoch is dead wrong here.. Actually a DAW is copying around more in memory than it‘s computing. It‘s easy to underestimate how many internal audio streams even a simple 16 channel setup causes. Of course it always depends on where your bottleneck is but yes, 150% faster memory can make your DAW perform 150% faster.

And no, we don‘t have quantum computes yet, computers still use timers and yes, they‘re important for how fast stuff gets around.

[Voyager]

Memory speed is very much important for a DAWs speed, Enoch is dead wrong here.. Actually a DAW is copying around more in memory than it‘s computing. It‘s easy to underestimate how many internal audio streams even a simple 16 channel setup causes. Of course it always depends on where your bottleneck is but yes, 150% faster memory can make your DAW perform 150% faster.

And no, we don‘t have quantum computes yet, computers still use timers and yes, they‘re important for how fast stuff gets around.

ok, i understand that speed memory is not what all matter, to get a fast and stable machine we have to make sure all components work well togheter and are on total symbiosis. Installing fast memory on a inadequate cpu and machine is like installing the best slick tires on a daily drive volkswagen, you won't get the best potential of the tires and not go "that" fast..

Now about frequency speed and timings i understand that you'll get faster memory and better bandwidth with 2133 cl9 than 1600 cl7 but cl9 has higher timings than cl7.

So is frequency speed or low timing latency more favorable for Reason ? or maybe frequency speed and low timing isn't something to compare because what's matter at the end is the overall memory speed and bandwidth ?

[normen]

ok, i understand that speed memory is not what all matter, to get a fast and stable machine we have to make sure all components work well togheter and are on total symbiosis. Installing fast memory on a inadequate cpu and machine is like installing the best slick tires on a daily drive volkswagen, you won't get the best potential of the tires and not go "that" fast..

Now about frequency speed and timings i understand that you'll get faster memory and better bandwidth with 2133 cl9 than 1600 cl7 but cl9 has higher timings than cl7.

So is frequency speed or low timing latency more favorable for Reason ? or maybe frequency speed and low timing isn't something to compare because what's matter at the end is the overall memory speed and bandwidth ?

The speed (i.e. overall throughput) is probably more important but it's hard to say without sifting through every single protocol involved. Theres probably ratios at certain points where the importance moves from speed to latency - but they don't have to be related to actual audio latency really. In general I'm just saying what you can save on a GPU for an audio PC you should not completely invest only in the CPU. Audio very much is a balancing game as you use a computer for things it wasn't made for so it would be stupid to only look at the CPU.

[EnochLight]

Memory speed is very much important for a DAWs speed, Enoch is dead wrong here..

Reason is affected very little by memory CAS speed/timings. Clock speed, as I mentioned earlier, certainly can help. I would love to see your legitimate data to prove otherwise, because my own testing with various RAM modules offered no noticble difference in Reason’s performance.

[normen]

Clock speed, as I mentioned earlier, certainly can help. I would love to see your legitimate data to prove otherwise, because my own testing with various RAM modules offered no noticble difference in Reason’s performance.

Actually I was only referring to your first comment (which you now re-stated) that Reason is in general not dependent on memory speed, only CPU speed. But if your statement is legitimate data then my legitimate data is that I got a reverb plugin instance increase of 33% when going from 100 to 133MHz memory back when I still measured such things. But obviously the plural of anecdote isn't data. If your processor is your bottleneck then making it faster will make the whole process faster. But the more you MIX rather than PROCESS the more important your memory speed becomes. Remember a single channel easily means 8-16 internal "channels", i.e. copies of the data - for the sends, main bus, inserts etc etc.

[EnochLight]

Clock speed, as I mentioned earlier, certainly can help. I would love to see your legitimate data to prove otherwise, because my own testing with various RAM modules offered no noticble difference in Reason’s performance.

Actually I was only referring to your first comment (which you now re-stated) that Reason is in general not dependent on memory speed, only CPU speed. But if your statement is legitimate data then my legitimate data is that I got a reverb plugin instance increase of 33% when going from 100 to 133MHz memory back when I still measured such things. But obviously the plural of anecdote isn't data. If your processor is your bottleneck then making it faster will make the whole process faster. But the more you MIX rather than PROCESS the more important your memory speed becomes. Remember a single channel easily means 8-16 internal "channels", i.e. copies of the data - for the sends, main bus, inserts etc etc.

If you were referring to this:

It's negligible. Reason is mostly affected by CPU speed. That difference in RAM will be hardly noticeable in Reason, if at all. CPU speed and SSD speed have the biggest effect overall.

...then I stand by that statement, still. My own testing seems to indicate as much, as well as the very lengthy (and multiple) forum threads here with people sharing their own data.

And when I say negligible, I am not saying that there isn't a difference at all. It's just crystal clear that CPU speed brings the greatest gains when coming to Reason. But you already know this.

[normen]

Actually I was only referring to your first comment (which you now re-stated) that Reason is in general not dependent on memory speed, only CPU speed. But if your statement is legitimate data then my legitimate data is that I got a reverb plugin instance increase of 33% when going from 100 to 133MHz memory back when I still measured such things. But obviously the plural of anecdote isn't data. If your processor is your bottleneck then making it faster will make the whole process faster. But the more you MIX rather than PROCESS the more important your memory speed becomes. Remember a single channel easily means 8-16 internal "channels", i.e. copies of the data - for the sends, main bus, inserts etc etc.

If you were referring to this:

It's negligible. Reason is mostly affected by CPU speed. That difference in RAM will be hardly noticeable in Reason, if at all. CPU speed and SSD speed have the biggest effect overall.

...then I stand by that statement, still. My own testing seems to indicate as much, as well as the very lengthy (and multiple) forum threads here with people sharing their own data.

And when I say negligible, I am not saying that there isn't a difference at all. It's just crystal clear that CPU speed brings the greatest gains when coming to Reason. But you already know this.

Cool. Anyway given the actual question of OP and the fact that audio is one of the relatively few applications where faster memory makes sense I'd still call your answer wrong - so you now know this

[EnochLight]

Cool. Anyway given the actual question of OP and the fact that audio is one of the relatively few applications where faster memory makes sense I'd still call your answer wrong - so you now know this

[4filegate]

[TheNoog]

This difference isn't going to do shit in Reason. It might load 0.01 seconds faster but that's it.

[TheNoog]

Cool. Anyway given the actual question of OP and the fact that audio is one of the relatively few applications where faster memory makes sense I'd still call your answer wrong - so you now know this

Lol. No. But hey, since you claim it, how about backing it up with some real world evidence.