| Author
|
FL Producer 5 Cubase SX 2 comparison.
|
PsYx
IsraTrance Junior Member
Started Topics :
12
Posts :
339
Posted : Dec 22, 2004 22:44
|
|
I suggest to go 2 >irc.QUAKENET.org-> #ENGINETEST |
|
|
Spindrift
Spindrift
Started Topics :
33
Posts :
1560
Posted : Dec 22, 2004 22:48
|
If there was a forum on audiotest.batcave.net at the moment it of course be better to keep the discussion there.
Edit..
Not meant instead or IRC that could be good as well....I'll check in there later
But there still would be discussion as it always is in a communty, thats why it can be great to work together.
It's many producers hanging out here and much input on the subject.
There have to be discussion if you want to perform a test like this it seems.
And if you want help to get a test together that serves the interests of the community you have to be prepared for much discussion.
The main question still with the test is what do you want to find out.
I have one suggestion:
How does the sound quality compare in real world usage?
Someone have a better definition or meaning I don't mind hearing that.
Maybe I'm the only one thinking that is what they like to know.
But if not I be willing to try to make a precise set of test guidelines to answer that question if thats what is asked.
But I would like to have them discussed as well in public before the test should be carried out, since the tests will be carried out by and observed by the community.
Better to hear the complaints when you still can change them, and thats whats great about a community to have many views and some discussion about the subject to get better results.
 (``·.¸(``·.¸(``·.¸¸.·`´)¸.·`´)¸.·`´)
« .....www.ResonantEarth.com..... »
(¸.·`´(¸.·`´(¸.·`´``·.¸)``·.¸)``·.¸)
http://www.myspace.com/spindriftsounds
http://www.myspace.com/resonantearth |
|
|
|
H2O
IsraTrance Junior Member
Started Topics :
16
Posts :
352
Posted : Dec 22, 2004 22:49
|
Ok, this will be the official mirc channel of this test and forum(fatherkike.proboards43.com) will be this one, for users who are not online. Then we will post full test here on isratrance, only when it is realy finnished.
Official forum of ADcomm is not ready.
so
fatherkike.proboards43.com as board
and
irc.QUAKENET.org-> #ENGINETEST as mirc live chat
|
|
|
|
thockin
Started Topics :
1
Posts :
114
Posted : Dec 22, 2004 22:59
|
Undertow - you can email me the files and I'll put them up on my webspace: enginetest@hockin.org.
As for more varied testing - if we determine that the basic mixing "engine" differs, then teh rest will differ too. If we can determine that they are significantly the same in mixing, then the rest matters. |
|
|
|
thockin
Started Topics :
1
Posts :
114
Posted : Dec 22, 2004 23:01
|
All hosts are welcome, of course. I can personally set up FL3 FL4 and FL5, ACID 3, and probably Sonar 3. I'll post the projects
|
|
|
|
Dreamthief
Started Topics :
3
Posts :
47
Posted : Dec 22, 2004 23:25
|
This test idea is good, but if everyone renders mp3 files of the results, it won't be a true representation.
I propose that test files used to test be 24bit aiff/wav just to keep things as standard as possible. White noise samples are a good idea, as they should represent the total freq band.
I will also check out the irc channel when I get home. It's 8am here now, and I should be home at about 4pm. If anyone is online, we can chat there.
Thockins parameters seemed pretty cool, if I think of anything more, I will add them.
|
|
|
|
thockin
Started Topics :
1
Posts :
114
Posted : Dec 22, 2004 23:55
|
|
UnderTow
Started Topics :
9
Posts :
1448
Posted : Dec 23, 2004 00:09
|
I'll resume things here:
I created 12 stereo 44.1Khz/24 bit files. Together they sum at exactly 0.0dBFS in Sonar 4. (And they might even sound musical to your ears  .
They are available at Tim's website at the URL he gave: http://www.hockin.org/~thockin/engine_test/
Anyony can download these files and do their own tests.
In the end we want to do tests with various hosts with the same settings (where possible) and compare results. We will use the guidelines posted by Tim earlyer in this thread. I think H2O will put the exact guidelines and test procedures on his site when it is ready. (Right?)
Discussions can be done on the forum that H2O posted and on the IRC channel and server that PsYx posted.
In the meantime, have fun with the files. 
UnderTow
|
|
|
|
H2O
IsraTrance Junior Member
Started Topics :
16
Posts :
352
Posted : Dec 23, 2004 00:12
|
BPM is 148.
When the test will be done we will post it on site at sequencers chart. |
|
|
|
thockin
Started Topics :
1
Posts :
114
Posted : Dec 23, 2004 00:23
|
|
Also, for the record - we will not exclude any host, if possible, and we will perform as much analysis of the results as possible. I want this to be *the* test people point to when someone claims X is better than Y. |
|
|
|
H2O
IsraTrance Junior Member
Started Topics :
16
Posts :
352
Posted : Dec 23, 2004 00:39
|
FL do not export 24bit. So we need to export to 16bit.
To set FL Producer 0dB follow this picture.
http://audiotests.batcave.net/3231.jpg
this should make FL Producer load samples as original.
Panning laws off and maximum volume at patern sequencer.
|
|
|
|
thockin
Started Topics :
1
Posts :
114
Posted : Dec 23, 2004 00:54
|
Sigh.
First, FL exports at 32 bit, which can be converted to 24 bit beautifully. Every other app uses 32 bit internally and converts to 24 bit on export.
Second, don't fiddle with the panning law. Just set the volume to max on all channels. When you are done, NORMALIZE THE RESULT before converting bitdepth.
|
|
|
|
H2O
IsraTrance Junior Member
Started Topics :
16
Posts :
352
Posted : Dec 23, 2004 01:59
|
|
Thockin, do as i said export then do as you said, compare the signal and you will have zero difference. |
|
|
|
thockin
Started Topics :
1
Posts :
114
Posted : Dec 23, 2004 02:03
|
|
yes, but my point is that you need to normalize the end result anyway for the test to be fully valid, so there is ZERO point in fiddling inside the host. On the contrary, when we get to panning tests, you *want* the host's normal behavior. |
|
|
|
thockin
Started Topics :
1
Posts :
114
Posted : Dec 23, 2004 11:40
|
I have started on 2 tests. The first simply validates that what you put in is what you get out.
Process:
1. Load the "Loops" tracks into the sequencer.
2. Set the track's volume to max.
3. Trigger the track once, on the first sample of the project (default velocity).
4. Render to the highest bit-depth and highest quality you can.
5. Open each result file in a wave editor (I used SoundForge 6.0).
6. Remove any extra file properties, such as project name or loop points. Save.
7. Normalize to -0.0 dB. Save.
8. Convert the bit-depth to 24 bit integer, if it is not already.
9. Compare.
The results on FLS are surprising - they are, in fact, different. I can't explain it, and I have asked teh FL peopel for input. More later.
The second test loads all 12 tracks up (nice tune, by the way, UnderTow).
Process:
1. Load all 12 tracks into the sequencer.
2. Set each track's volume to max.
3. Trigger each track once, on the first sample of the project (default velocity).
4. Render to the highest bit-depth and highest quality you can.
5. Open each result file in a wave editor (I used SoundForge 6.0).
6. Remove any extra file properties, such as project name or loop points. Save.
7. Normalize to -0.0 dB. Save.
8. Convert the bit-depth to 24 bit integer, if it is not already.
9. Compare. |
|
|
|
