EDIT_please ignore
Last edited by Trakstar on Wed Jul 04, 2012 8:07 am; edited 1 time in total

And I apologize for being a dick, nothing meant by it just found it funny, get a few freq analysis read outs too next time. At least you tried, and people who never get things wrong(or think they do) can be proper assholes. I think its Hink that puts it in his sign "I never learned anything from being right".
Try again and see what happens with a few other diagrams for preference or something.

Ok. Such clear null test is not very useful, i think. At least for me:-) I hear that reaper is not as good at low end as samplitude, But I work in reaper, because it's more convenient for me than samplitude)).

How can this test be correct in practise aspect if it has such restrictions - "do not insert this pluggin during the test! Ohh - ok, this pluggin you can use in test! You do not like this VST? Ok, but with you favorite VST this test will be incorrect!" Ha ha))

It's a fake, not a test. IMHO!

Because if you test a plugin that has a random element to it you wouldn't test if the host sounds different,they would by default. If you really want to know if two hosts sounds different you need plugins you know will produce the exact same results every time. Otherwise you're just testing if there's a difference between two instances of the same plugin and we already know they will not sound the same because of the random element.

EDIT: In the same way you would use the exact same audio snippet in both hosts,not two different ones.

Is there any other kind ?

REAPER is exactly the same at low end as samplitude. it has been proven numerous times and can be proven million times more.

this test is correct in practice because believe it or not, computer audio is very deterministic. if one plugin works and sums correctly, they all do. to host, sound is just numbers, it doesn't care whether these numbers come from EQ or compressor, or whether there's two numbers to add, or two million numbers to add. provided the numbers are always the same (i.e. there's no inherent randomness), the output will always be exactly the same, just like adding 2 and 3 will always give you 5, no matter how many times you count.

the test stops being correct and valid when a non-deterministic value is introduced. if you put two EZDrummer tracks side by side and give them exact same MIDI file, they will produce different output. even in the same host. try it, you'll be surprised. that is happening because EZDrummer itself is the non-deterministic element - it does "humanization" and it provides several samples per on note, and chooses randomly between them. The very thing that avoids famous "machine gun effect", renders the null test invalid because the audio is different every time.

therefore, to truly test the *host* and not the plugins, you have to carefully choose only those plugins which behave deterministically and produce exact same output every time. just like you would pick the same WAV sample to test hosts' summing, not slightly different ones.

I just made a simple loop from one shots. Rendered it to One stereo file.
Thеn place this stereo file in reaper again and rendered again. Then i made this job with samplitude. No dither in reaper and samplitude.
16 bit, 44100.
So, Please, tell me why rendered from reaper is louder then from samplitude? What is wrong?

http://www.filedropper.com/mix

no one can tell you why without you posting source files and projects.

ok. Just a moment

Probably your pan law settings are different.

i am downloading samplitude demo now. which VST's are you using in the test?

No VST at all. Samplitude Pro-X and Reaper 4

Here is the project.
http://www.filedropper.com/showdownload.php/4kvr

I really will be happy if you could tell me what I'm doing wrong in this case.

Thanx.
Last edited by Pupok on Wed Jul 04, 2012 9:10 am; edited 1 time in total

i certainly don't see anything unusual in REAPER 4 project. waiting for Samp Demo to download...

Ok..

And im still listening so dont mind me