The test isn't finished yet. You are acting as emotionally as the rest of us thread cops.

It would be interesting to know how close they came to nulling, and if rewire was involved.

ORIGINAL: semieval665
THe flames for asking a question..........man.........how "little" do you have to be to do that?

Most flames regarding people's questions happen because the answers are easily found in the Sonar manual or elsewhere on this forum by using the "Search" function. There are rare exceptions, but they are very rare.

ORIGINAL: Ron Kuper [Cakewalk]

I only just got the software I need to run shea's test file yesterday. (His file consisted of Reason 2.5 tracks and Kontakt tracks.)

I did an A/B in a "pan-law adjusted" build of SONAR (S4 will have pan law) against Nuendo. To my ear, the two mixes sounded the same. However, mixes exported from SONAR and Nuendo didn't null out. I need to dig deeper to see why. I was going to isolate individual tracks to see if it was somehow track specific.

Ron...I have to say you have been quite patient and extremely detailed in your approach to this problem, I would imagine most other companies would either not discuss or reveal the differences that you mentioned below for fear of appearing to have an inferior product.

"To my ear, the two mixes sounded the same. However, mixes exported from SONAR and Nuendo didn't null out. I need to dig deeper to see why. I was going to isolate individual tracks to see if it was somehow track specific."

Obviously there is nothing inferior about Sonar and thats why I use it.It is far more stable than Nuendo from what I hear and far easier to pick up. Featurewise...even after almost a year of usage I am still discovering new and useful features.Great.

Also apart from the myriad of other Cakewalk related duties you have to attend to, offering to help solve this issue and give so much time to answer the original question is a real example of customer support. I am very impressed Ron. Very pleased, keep it up.

ORIGINAL: Ron Kuper [Cakewalk]
I did an A/B in a "pan-law adjusted" build of SONAR (S4 will have pan law) against Nuendo. To my ear, the two mixes sounded the same. However, mixes exported from SONAR and Nuendo didn't null out. I need to dig deeper to see why. I was going to isolate individual tracks to see if it was somehow track specific.

As far as nulling, would dither have anything to do with non-nulling? Since dither is random noise, then even if all other things are equal, you would still have that very low level random noise that would cause the files to not null out. Have you been doing the tests with dither off?

When the files were differenced, what was the general amplitude of the difference file?

did you check that the mixes started at exactly the same time?

They didn't, but I slid the waves over to compensate for that.

As far as nulling, would dither have anything to do with non-nulling? ...
When the files were differenced, what was the general amplitude of the difference file?

In this case dither was definitely not the cause. The differences were pretty significant, somewhere in the audible phasing range.

are the difference concentrated in certain bands of the spectrum?

No.

Oh al, you are the best.............!!!!!!!!!!!!!!!!!!!

On my main machine i have everything on it including Nuendo.
On the other machine i just have reason and kontakt.
When I Play the file in both sonar and nuendo on my main machine using
rewire and soft synth the normal way, they sound different to me.
But when i play them both again using the same sounds from the other
machine via midi they sound identical to me.

Shea, why didnt you do this test on both systems right from the beginning.

ORIGINAL: alexniedt

ORIGINAL: semieval665
THe flames for asking a question..........man.........how "little" do you have to be to do that?

Most flames regarding people's questions happen because the answers are easily found in the Sonar manual or elsewhere on this forum by using the "Search" function. There are rare exceptions, but they are very rare.

in

Whatever dude

go back and actually READ the posts. Get an english tutor if you don't understand things.

ORIGINAL: mark s

silly me: I was hoping this thread was dead. I'll be waiting for news from Ron

I couldn't believe it's still going, I can't believe I'm writing this..............

Must be board.

shea,
thanks for finally getting around to delivering an actual test of what you're talking about.

And thanks again to Ron for your supernatural patience and dilligence and clear devotion to not only making cakewalk the best you can, but also to letting your customers in on the process. This kind of candor and dedication is rare in executives of companies of any stripe, and recording software is no exception. I hope you get paid fairly for your excellent work.

This may yet turn into a very constructive and informative thread. Now if we could just delete the first 19 pages of ranting, invective, and flame wars...

Cheers.

ORIGINAL: semieval665

ORIGINAL: alexniedt

ORIGINAL: semieval665
THe flames for asking a question..........man.........how "little" do you have to be to do that?

Most flames regarding people's questions happen because the answers are easily found in the Sonar manual or elsewhere on this forum by using the "Search" function. There are rare exceptions, but they are very rare.

in

Whatever dude

go back and actually READ the posts. Get an english tutor if you don't understand things.

Wow, alright DUDE...how much of a jerk can you possibly be? Your comment wasn't the least bit specific, and I thought you were just whining about the forum in general.

I've yet to read one productive post from you. I'm with Al. To hell with you...

ORIGINAL: shea
Version 1.5 and 2.5 latest updates

Did you happen to get a chance to... uh... well... ok I'll say it: do #2, that I suggested above?

Seriously, just mute each MIDI track in turn on Sonar to see if anything pops out as obvious...

[EDIT: oops, I meant "solo each" not "mute each"]

- Keith

Steve,

Having read your post several times trying to understand what you are trying to say I believe there are some gaps in your understanding of the techniques you discuss. Firstly, rounding and truncating are pretty simple and very well defined steps. Whatever the algorithm used the result will be either right or wrong. You would need a seriously deficient programmer to not get either of these functions correct - even if programming in machine code.

The rest of your speculation can be summed up in one well known term - dither.

Now we know that the Sonar dither has been criticised in the past and various dither algorithms and noise shaping have been examined extensively. I presume this is what you refer to when you say "what algorithmic shape to cut it with". This has nothing to do with rounding or truncating as you suggest but is an alternative to either of these functions in an attempt to hide the very low level digital noise it is claimed the simple process of rounding creates. (I do not think anyone in their right mind would try to use truncation to create the 24 bit audio from 32 bit float).

I will agree that there may be differences using different dither algorithms and so does Ron Kuper but the tests being conducted are carefully not using dither to eliminate this as a possible cause. This is precisely why Ron has been able to completely null various tests. I do not believe there is any magic going on here. My personal belief is that the undoubted sonic differences being heard in many tests could be down to timing issues. I have often been surprised at what sonic effect can be caused by very minor timing changes in a complex mix and I believe that some of the relatively large differences heard by Shea and noted by Ron Kuper in his nulling tests could be of this sort of magnitude due more to timing issues rather than some other well defined and very low level algorithmic difference. Interestingly the seemingly large level differences that Ron has noted have been produced ( I strongly presume) without the use of dither.

I think that there is something in all of this thread that needs to nailed down but I really do not think that the suggestions you are making are really landing in the same ball park.

I am looking forward to Ron's next post. If my own speculation turns out to be right it could be a difficult one for Ron to address. Then a again it could be that Sonars timing is the tight one and poorer timing could just sound nicer. Has anyone even considered timing dither??? I look forward to hearing the true answers.

Edit. When I am referring to timing issues I am not just thinking of things like the sample accurate recording being discussed in other threads but also possible variations in timing during the mixing process. I personally think this is more significant than possible sonic differences between rounding/truncating/dither processes.

Edit 2. Although I am fairly sure that timing is not an issue with straght recording (other than latency, being discussed elsewhere) I think the possible errors I am referring to are more likely when using Dx(i), VST(i) plug ins as in this case.

Steinberg is crap in my opinion. I used some of their products in the past and had:
Multiple crashes
Terrible tech support
Zero updates to resolve conflicts

Would I buy anything Steinberg?
Never again...ever
Because for me it's crap in... crap out...
All things being equal, I'm sure you can find differences in every product, but from the listener standpoint most people have no idea of what you are talking about and really don't care.
Deafinity

once again..

there was/is no reason that this thread went on that long..

it unbelievable, there is still less than 3 posts that actually have some infos about the topic!