sx perfomance

[ericj23]

we have all heard of those nice craxors who reckon that cubase would improve in performance if they took away the copy protection. Being a cubase user i thought i would investigate this so i set up a demo thing in moobase - three channels of audio a few vsts from komplete - a smattering of camel audio demo fx

i then set exactly the same set up in the demos of fruity loops, tracktion 2, samplitude & I would ahve done sonar if it came with the fxpansion wrapper ! I then used the xp cpu useage meter as the song looped around. The plan is if cubase has the alleged 50% perfomance hit this should be nice and obvious

results
cubase cpu 35-41 % memory 455 MB

fruity cpu 35-41 % memory 177 MB

samplitude 35-41 % memory 357 MB

tracktion 32-45 % memory 371 MB

so cpu useage is pretty much the same - there was a lot of fluctuation and i cant realistically say whether they were completely the same - certainly the very occasional lower numbers for traktion should be contrasted with the slightly higher peaks -it certainly was mainly between 37-39% as the others

memory useage is slightly worse in cubase - but all the tested sequencers are much worse than fruity - is this something to do with low buffers ??

but anyhoo the question at hand

50% loss in performance

or

the craxors are liars
who i guess want there next crack to be easier


after doing this I am certainly not changing host - and while cubase makes lots of dongle calls this will be for opening editors - midi editing etc- ie they do not appear to affect playback performance to a great extent

i have no issues with a slightly delay in opening a window while the dongle is called - it still happens pretty much instantly as far as i can see

so can we put this myth to bed please -

PS if anyone wants the cubase song I can happily mail it to em -alas i couldn't save the song in other progs(except tracktion) but i can send the midi files used (although the audio came from my copy of phatfactory)

[uselessmind]

How does this comparison proof that the cp doesn't effect the performance of Cubase?
You can do all kinds of tests,doesn't matter if you don't have a Cubase version without cp.

We can believe either the crackers,or...

Well if Steinberg would clearly say its bullshit we could believe them.
I would.
But they didn't say a thing about it in their last statement.

However,i still fail to understand why this is so important?
I wouldn't care why a software uses too much cpu/whatever.I would want to have it fixed,i don't care how.

[jasonsantiago]

1) This doesn't disprove anything. Cubase could be more efficient then all of those hosts and we'd never know.

2) A better test would be to run the same plug-ins in SX 1, 2, and 3 and see the results. Particularly between 2 and 3, since that's where most of the big dongle call changes are noticed from port sniffing.

Too bad steinberg hasn't addressed this at all, and probably will not. The crackers alleged 20-40%, so 50% is a bit high.

[bduffy]

I just don't see why a dongle has to be called with each and every action - why is that necessary? So what if I decide - for some reason - to pull my dongle during mixing?...am I missing something? What's the point?

[ericj23]

1) This doesn't disprove anything. Cubase could be more efficient then all of those hosts and we'd never know.

2) A better test would be to run the same plug-ins in SX 1, 2, and 3 and see the results. Particularly between 2 and 3, since that's where most of the big dongle call changes are noticed from port sniffing.

Too bad steinberg hasn't addressed this at all, and probably will not. The crackers alleged 20-40%, so 50% is a bit high.

OK so youre telling me that cubase is 20% more efficient than 3 other hosts (which are all the same) but its lost under dongle calls

or

the dongle calls make absolutely no difference on playback becuase they are only made when opening editors - moving midi data etc

and whats the point you ask ? - thats just the route of another warez argument - the copy protection stays - and steinberg have already stated this

the point of the little test was to show (and as far as i am concerned i have) that any alleged performance problem caused by the cubase copy protection is clearly very small (it certainly is not 20 -40% - its not even 1 percent !)

[cold c]

2) A better test would be to run the same plug-ins in SX 1, 2, and 3 and see the results. Particularly between 2 and 3, since that's where most of the big dongle call changes are noticed from port sniffing.

Too bad steinberg hasn't addressed this at all, and probably will not. The crackers alleged 20-40%, so 50% is a bit high.

Try it for yourself with the official versions of SX2 and SX3 using the same project, there is no difference in CPU utilisation in task manager, no more or less pops/clicks at all.

I don't doubt that 20-40% of the application commands are wrapped up in encryption and CP, but what has this got to do with playback performance?

It could have been the reason behind early bugs (in earlier versions) such as a slow right click menu pop-up, but this is not project playback performance.

And yes, I would rather SX3 did not require a dongle, but this is because of Steinberg's lost dongle policy, and the fundamentals of 'pure' native software with the least dependance on an unnecessary piece of hardware (for example: whereas software can be backed-up as many enterprise environments do, a hardware dongle can't and is therefore an added 'risk').

[loomchild]

A little off-topic but the differences in mem consumtion are pretty interesting.

[bduffy]

A little off-topic but the differences in mem consumtion are pretty interesting.

Yeah, I'm not crazy about it, kinda embarrasing. But it's no secret that SX opens with a lot of resources. Don't know why or if it's necessarily bad.

[ericj23]

what i was more amazed by was how differnt fruity was - it is a lot less consuming

is this becasue PDC needs lots of buffering of outputs ?

[original flipper]

HI

Especially if your running on 512meg!

Flipper.

[fizbin]

Well if Steinberg would clearly say its bullshit we could believe them.
I would.
But they didn't say a thing about it in their last statement.

Official Steinberg statement -
http://forum.cubase.net/phpbb2/viewtopic.php?t=18196

fizbin

[parky]

edit.

[jasonsantiago]

The official steinberg statement still doesn't address the "encryption may take up cpu resources" issue.

OK so youre telling me that cubase is 20% more efficient than 3 other hosts (which are all the same) but its lost under dongle calls

or

the dongle calls make absolutely no difference on playback becuase they are only made when opening editors - moving midi data etc

no, neither of those scenarios are what I am saying. I am saying since we don't know how efficient cubase would be without the encryption, simply comparing them to other programs is not completely fair, although it can be used as one piece of evidence to gain a general, albeit speculative picture (all these results are anyway). I think the SX2 and SX3 test would be a better test, since in this case you have a cubase build without the super heavy encryption, and then of course the one with it. As one user states, he noticed 0 difference. This re-inforces what you are saying. However Steinberg not broaching this subject makes me still paranoid.

I am going to try this myself and install SX2.

For the user that did try the SX2 and 3 comparison, did you check memory and hard drive usage as well (just to be thorough)?

[parky]

I just don't see why a dongle has to be called with each and every action - why is that necessary?

Rampant paranoia on Steinberg's part?

[ericj23]

the flaw with that is that sx 2 had super heavy encryption too

that one took the crackers 6 months or so - this one a year - but does this extra cracking time does mean that the protection is more resource draining ?

there is nothing to suggest it does

i just did the test to check that the grass wasn't greener - it's not

anyhoo this is all a sure sign i am very bored - i'm off to do soemthing