Behringer ADA8000, first impressions
Hi Stige,
I'm intrigued by your experiment. Some questions first, then I'll tell you why I ask.
What generated the 1k sine wave? Was it a balanced or an unbalanced source?
Did the 1k sine wave go through the Mindprint and Behringer at mic' level?
If it was at line level, did you try it directly into Pulsar II and is your II a Classic, i.e. unbalanced? If so, how did you go from balanced to unbalanced (assuming your source was balanced)? If the source was unbalanced, how did you go into the Behringer?
In Samplerate Settings, when you recorded the ADA8000, did you select "Slave" and then ADAT A (or B)?
When you recorded through the Pulsar II analog, did you change the Samplerate Settings to "Master?"
I definitely see a difference, but I'm not convinced the difference you hear is due to the A/D converters. Here's why.
In my system, I can listen to signals converted by a Pulsar II plus, an RME ADI-8 Pro and an RME ADI-AE, and frankly, they all sound good to me, even though the ADI-8 Pro has a multi-dB advantage on the Pulsar II Plus and even the RME AE. I can see the difference, I can hear some difference, but those are more like flavors to me. One thing, though, all those devices in my system are controlled by an external clock. The syncplate is a must for multiple digital sources.
I don't doubt you hear a difference, but if you went through the Mindprint, I would imagine that raggedness you're seeing from the Pulsar II is due to a flaky tube.
That's assuming you changed the clock to allow the Pulsar II to be master when it had its 'turn in the barrel.' If not, then that's a culprit. Get a syncplate.
If you changed clock for each experiment, then the smoothness of the Behringer ADA8000 may be testimony to the quality of its clock more than its AD converters.
Thanks for doing the experiment. I think I'll learn some interesting things from it once I find out a few more details.
john
<font size=-1>[ This Message was edited by: jabney on 2004-02-15 22:21 ]</font>
<font size=-1>[ This Message was edited by: jabney on 2004-02-15 22:35 ]</font>
I'm intrigued by your experiment. Some questions first, then I'll tell you why I ask.
What generated the 1k sine wave? Was it a balanced or an unbalanced source?
Did the 1k sine wave go through the Mindprint and Behringer at mic' level?
If it was at line level, did you try it directly into Pulsar II and is your II a Classic, i.e. unbalanced? If so, how did you go from balanced to unbalanced (assuming your source was balanced)? If the source was unbalanced, how did you go into the Behringer?
In Samplerate Settings, when you recorded the ADA8000, did you select "Slave" and then ADAT A (or B)?
When you recorded through the Pulsar II analog, did you change the Samplerate Settings to "Master?"
I definitely see a difference, but I'm not convinced the difference you hear is due to the A/D converters. Here's why.
In my system, I can listen to signals converted by a Pulsar II plus, an RME ADI-8 Pro and an RME ADI-AE, and frankly, they all sound good to me, even though the ADI-8 Pro has a multi-dB advantage on the Pulsar II Plus and even the RME AE. I can see the difference, I can hear some difference, but those are more like flavors to me. One thing, though, all those devices in my system are controlled by an external clock. The syncplate is a must for multiple digital sources.
I don't doubt you hear a difference, but if you went through the Mindprint, I would imagine that raggedness you're seeing from the Pulsar II is due to a flaky tube.
That's assuming you changed the clock to allow the Pulsar II to be master when it had its 'turn in the barrel.' If not, then that's a culprit. Get a syncplate.
If you changed clock for each experiment, then the smoothness of the Behringer ADA8000 may be testimony to the quality of its clock more than its AD converters.
Thanks for doing the experiment. I think I'll learn some interesting things from it once I find out a few more details.
john
<font size=-1>[ This Message was edited by: jabney on 2004-02-15 22:21 ]</font>
<font size=-1>[ This Message was edited by: jabney on 2004-02-15 22:35 ]</font>
Hi John
I'll try give a better explanation how I tested. My test wasn't 'scientifically' correct. But it was a 'real world' test that could happen in any time in practice.
I have a Pulsar II with z-link plate (unbalanced) + Pulsar xtc
I made a sine 1KHz in soundforge with linear fade-in, peak -0.1dB. Then I fed it through Pulsar spdif out to the minidisc recorder. Then I trimmed the volume down (in analog domain) so that both ADA8000 and Pulsar converters reached -60dB at the peak of the sine. All connetions to the both converters were unbalanced. I kept ADA8000 as master unit all the time. It's strange, but I have a feeling that STM4896 mixing sounds better with ADA8000 as master. That must be bullsh!t though in the light of technical facts.
In practice, I had to feed the signal much lower volume to the pulsar input than to the ADA8000 input. This means also, Pulsar picked up more noise/interference from the device where it was connected to. This is why pulsar's figure seemed more distorted. But this is a real world example, as I need to feed everything at very low level into pulsar AD to avoid clipping. With ADA8000 I can use louder input signals and at the same time, the noise floor stays lower. I know, pulsar with balanced inputs would solve all these noise problems, but I thought I would be using external converters anyway. I agree, sync plate would be a must. Hope this gave you better image of the test.
I'll try give a better explanation how I tested. My test wasn't 'scientifically' correct. But it was a 'real world' test that could happen in any time in practice.
I have a Pulsar II with z-link plate (unbalanced) + Pulsar xtc
I made a sine 1KHz in soundforge with linear fade-in, peak -0.1dB. Then I fed it through Pulsar spdif out to the minidisc recorder. Then I trimmed the volume down (in analog domain) so that both ADA8000 and Pulsar converters reached -60dB at the peak of the sine. All connetions to the both converters were unbalanced. I kept ADA8000 as master unit all the time. It's strange, but I have a feeling that STM4896 mixing sounds better with ADA8000 as master. That must be bullsh!t though in the light of technical facts.
In practice, I had to feed the signal much lower volume to the pulsar input than to the ADA8000 input. This means also, Pulsar picked up more noise/interference from the device where it was connected to. This is why pulsar's figure seemed more distorted. But this is a real world example, as I need to feed everything at very low level into pulsar AD to avoid clipping. With ADA8000 I can use louder input signals and at the same time, the noise floor stays lower. I know, pulsar with balanced inputs would solve all these noise problems, but I thought I would be using external converters anyway. I agree, sync plate would be a must. Hope this gave you better image of the test.
Thanks Stige,
That gives me a much clearer picture of the test method. If you get curious, here are a few variables you may wish to try:
1. Keep a constant level from the minidisc, and adjust the gain on the Behringer ADA8000 to match the Pulsar II as measured at the sfp mixer. Then run your test.
2. Having done the above, change the clock to have the Pulsar II act as master and the Behringer as slave (from reading the ADA8000 manual, it looks as if you can send it clock via a lightpipe cable from the Pulsar II to the ADAT in of the ADA8000 and switching its SYNC to ADAT in. In sfp, just route any old signal to a pad on the Pulsar II ADAT Out module).
3. If that's not enough
try burning a CD with your 1k signal to bypass any possible ATRAC artifacts.
Again, thanks for the heads-up on the ADA8000. Looks to be pretty useful.
john
That gives me a much clearer picture of the test method. If you get curious, here are a few variables you may wish to try:
1. Keep a constant level from the minidisc, and adjust the gain on the Behringer ADA8000 to match the Pulsar II as measured at the sfp mixer. Then run your test.
2. Having done the above, change the clock to have the Pulsar II act as master and the Behringer as slave (from reading the ADA8000 manual, it looks as if you can send it clock via a lightpipe cable from the Pulsar II to the ADAT in of the ADA8000 and switching its SYNC to ADAT in. In sfp, just route any old signal to a pad on the Pulsar II ADAT Out module).
3. If that's not enough
try burning a CD with your 1k signal to bypass any possible ATRAC artifacts.Again, thanks for the heads-up on the ADA8000. Looks to be pretty useful.
john
Hi, I did another test with setting ADA8000 input gain equal to pulsar input. Now they both seem to be quite similar, what comes to picking up noise/interference. But still ADA8000 gives a bit lower noise floor. If I disconnect both converters and keeping the gain still equal to pulsar, the noise floor difference is greater (ADA8000=106dB, Pulsar=96dB)
My only option just at the moment is to keep ADA8000 as master, because I'm waiting my lightpipe order to arrive. I can't use ADA8000 outputs at the moment, neither set it to slave.
MindPrint is ready to bin, burns the fuse at the very moment when switching on
<font size=-1>[ This Message was edited by: Stige on 2004-02-17 09:59 ]</font>
My only option just at the moment is to keep ADA8000 as master, because I'm waiting my lightpipe order to arrive. I can't use ADA8000 outputs at the moment, neither set it to slave.
MindPrint is ready to bin, burns the fuse at the very moment when switching on
<font size=-1>[ This Message was edited by: Stige on 2004-02-17 09:59 ]</font>
Stige, if it's a failure of the powersupply there might be a chance to service it at a regular TV shop (those with trained staff - not the supermarkets).
I have another suggestions for some more beefy tests:
how about some transients (in various frequency ranges) provided by KickMe ?
I bet this will be less boring than just a different noise floor
cheers, Tom
I have another suggestions for some more beefy tests:
how about some transients (in various frequency ranges) provided by KickMe ?
I bet this will be less boring than just a different noise floor
cheers, Tom
Hi Peeps,
I just got my ADA8000, so I thought I could also participate to the first impression...
I have an SX project where I have several track of a the same pattern from my roland TD-8 (there are drum, percussion, bass, keys, etc ..) as a test base.
I synched (w/ midi clock) each track and record the exact same sequence with the different I/O I have :
1) PulsarII (x2), LunaII, creamware I/O box, Fostex VC-8, and now ADA-8000 ...
I set up the ADA8000 (as slave) and started recording the same patern..
I was not very convinced right away, and my fear was confirmed when I started to compare the ADA8000 track to the I/O box track.
(all tracks have been DCed and normalized).
It sounded indeed clear/bright, but thin, cold, without punch ... better than the fostex though, but not to replace any Creamware converters I have...
Then I don't know why, but started another test right away in the same condition, but this time I put the ADA8000 as Master, and started recording ....
Well, what a difference !!??!? It had more punch, it became warmer, the basses were better, the whole definition of the sound in general...
what a relief ! I even don't know which one I like best now
But now the ADA8000 is a good alternative.
Why such a difference ? Jitter ?
<font size=-1>[ This Message was edited by: proximo on 2004-02-24 14:40 ]</font>
<font size=-1>[ This Message was edited by: proximo on 2004-02-24 17:37 ]</font>
I just got my ADA8000, so I thought I could also participate to the first impression...
I have an SX project where I have several track of a the same pattern from my roland TD-8 (there are drum, percussion, bass, keys, etc ..) as a test base.
I synched (w/ midi clock) each track and record the exact same sequence with the different I/O I have :
1) PulsarII (x2), LunaII, creamware I/O box, Fostex VC-8, and now ADA-8000 ...
I set up the ADA8000 (as slave) and started recording the same patern..
I was not very convinced right away, and my fear was confirmed when I started to compare the ADA8000 track to the I/O box track.
(all tracks have been DCed and normalized).
It sounded indeed clear/bright, but thin, cold, without punch ... better than the fostex though, but not to replace any Creamware converters I have...
Then I don't know why, but started another test right away in the same condition, but this time I put the ADA8000 as Master, and started recording ....
Well, what a difference !!??!? It had more punch, it became warmer, the basses were better, the whole definition of the sound in general...
what a relief ! I even don't know which one I like best now
But now the ADA8000 is a good alternative.
Why such a difference ? Jitter ?
<font size=-1>[ This Message was edited by: proximo on 2004-02-24 14:40 ]</font>
<font size=-1>[ This Message was edited by: proximo on 2004-02-24 17:37 ]</font>
Thanks for the report proximo...i've just order myself an ADA8000 (to go with the Pulsar 1+ I just got) and I was a bit worried when I read the first part of you report!
I'll make the same comparison when I get mine, but it sounds like the ADA8000 works best a master (I guess it has a decent clock?). Are you using a word clock connection at all, or is it just clocking through the ADAT connection?
cheers.
I'll make the same comparison when I get mine, but it sounds like the ADA8000 works best a master (I guess it has a decent clock?). Are you using a word clock connection at all, or is it just clocking through the ADAT connection?
cheers.
Perhaps you could test using pulsar slaved to ADA8000 and then mixing with STM mixers. Do you hear any difference?
Proximo, I have a bit different experience. I recorded a bass guitar track using ADA8000 as slave. I noticed the sound was very detailed and clearly defined low-end. I didn't test putting ADA8000 as a master, I was satisfied with the results.
I'm thinking now that all that improvement in the sound using ADA8000 whether it's master or slave, must be due to my faulty MindPrint, which corrupted my recordings slowly, letting me wondering why everything sounds like sh!t.
Proximo, I have a bit different experience. I recorded a bass guitar track using ADA8000 as slave. I noticed the sound was very detailed and clearly defined low-end. I didn't test putting ADA8000 as a master, I was satisfied with the results.
I'm thinking now that all that improvement in the sound using ADA8000 whether it's master or slave, must be due to my faulty MindPrint, which corrupted my recordings slowly, letting me wondering why everything sounds like sh!t.
Ok I re-listened to what I recorded yesterday, and I found the same result .. but wait ... I did tons of tests today, ie : recording the exact same sequence in master/slave and did not hear the same amount of difference, but I did re-arrange my toslink cable (I have no sync-plate), so is that an explanation ? .. have no idea.
And let me tell you .. today I went totally crazy with this test !
I realized that none of the recording I did sounded the same.
I mean even in the exact same configuration (master or slave), when I do, let's say 6 subsequent recording of the same sequence, on 6 different tracks, they all sound different, I mean different ?????
So I thought is the ADA8000 the guilty one ?
So I did the extact same test with the Creamware I/O BOX (SFP as Master) .... and I had the same result !
What the hell is going on here ?
At first you did not really hear it, but then I used my headphones (Sennheisser HD-600 which is very sensitive), and when I looped at a specific location of the sequence where there is a low-end bass, a kind of synth pink noise and a synth rezo with a closing VCF .... not 1 recording sound like the other NONE ?!
I am completely lost
FYI :
recording was made at 48kHz.
I used SX 2.0 with the ASIO1-FLT Dest 64 SFP driver, and the SX Project was set to 48kHz/24bit.
Anyway, for the people who still hesitate again about the ADA-8000 .. well don't and go for it, I have done a lot of comparaison the the creamware Box and it is at the same level .. if not better
(I did not test the Mic Pre-amp, as I used only the line level).
I will upload the small waves you hear the difference of recording later ...
I mean that's unbelievable...
<font size=-1>[ This Message was edited by: proximo on 2004-02-25 10:18 ]</font>
And let me tell you .. today I went totally crazy with this test !
I realized that none of the recording I did sounded the same.
I mean even in the exact same configuration (master or slave), when I do, let's say 6 subsequent recording of the same sequence, on 6 different tracks, they all sound different, I mean different ?????
So I thought is the ADA8000 the guilty one ?
So I did the extact same test with the Creamware I/O BOX (SFP as Master) .... and I had the same result !
What the hell is going on here ?
At first you did not really hear it, but then I used my headphones (Sennheisser HD-600 which is very sensitive), and when I looped at a specific location of the sequence where there is a low-end bass, a kind of synth pink noise and a synth rezo with a closing VCF .... not 1 recording sound like the other NONE ?!
I am completely lost
FYI :
recording was made at 48kHz.
I used SX 2.0 with the ASIO1-FLT Dest 64 SFP driver, and the SX Project was set to 48kHz/24bit.
Anyway, for the people who still hesitate again about the ADA-8000 .. well don't and go for it, I have done a lot of comparaison the the creamware Box and it is at the same level .. if not better
(I did not test the Mic Pre-amp, as I used only the line level).
I will upload the small waves you hear the difference of recording later ...
I mean that's unbelievable...
<font size=-1>[ This Message was edited by: proximo on 2004-02-25 10:18 ]</font>
proximo, I don't your setup, so I reverse the scenario starting from your results.
Assuming the track consists of several different sources which do have overlaying frequencies (of course - the normal situation) then if the timing drifts in the sub-milliseconds range you will have 'intermodulation' effects in certain parts of the spectrum.
Some frequencies get extinguished, some emphasized.
Since it's about very subtle time units, each recording can in fact sound different if you listen closely.
Does this make some sense regarding your setup ?
cheers, Tom
Assuming the track consists of several different sources which do have overlaying frequencies (of course - the normal situation) then if the timing drifts in the sub-milliseconds range you will have 'intermodulation' effects in certain parts of the spectrum.
Some frequencies get extinguished, some emphasized.
Since it's about very subtle time units, each recording can in fact sound different if you listen closely.
Does this make some sense regarding your setup ?
cheers, Tom
Hi Tom,Hi Stige, thanks for your replies.
@Astroman : The 'intermodulation' effects you describe seems to be exactly what I am having : ie depending on the track/recording the same sound seems to have some frequencies which have dissappeared ...
What I don't know, if it is normal in my setup/situation :
Hardware :
1)PulsarII
2)PulsarII+
3)LunaII
4)Luna I/O Box
5)ADA8000
Source of recording :
Only the same sequence (multi-timbral) coming from only 1 single synth (Korg Trinity).
Does is still make sense ?
Can I have the 'intermodulation' effects with this sitation too ?
@Stige
I will try with ASIO2, but I tried directly to record into the STS4000 (so without asio drivers), directly from the source and had the same problem ?
Cheers guys,
Proximo.
@Astroman : The 'intermodulation' effects you describe seems to be exactly what I am having : ie depending on the track/recording the same sound seems to have some frequencies which have dissappeared ...
What I don't know, if it is normal in my setup/situation :
Hardware :
1)PulsarII
2)PulsarII+
3)LunaII
4)Luna I/O Box
5)ADA8000
Source of recording :
Only the same sequence (multi-timbral) coming from only 1 single synth (Korg Trinity).
Does is still make sense ?
Can I have the 'intermodulation' effects with this sitation too ?
@Stige
I will try with ASIO2, but I tried directly to record into the STS4000 (so without asio drivers), directly from the source and had the same problem ?
Cheers guys,
Proximo.
For those that are interested in such things, you can read all about the chips that Behringer use in the ADA8000 here:
http://www.alesis-semi.com/products.htm
cheers.
http://www.alesis-semi.com/products.htm
cheers.
Hasn't anyone tested the mic pres in the AD8000? I already have an ultra 16, as many of you do it seems. I have a high end API mic pre,but it only has 4 channels. I need a 8 channel mic pre to record drums.With all the above gear plus more I am low on cash. I wonder how the AD8000 mic pres compare to say,a mackie,which I already have a vlz and a vlz pro. I could sell one to buy AD8000 if mic pres are better.
Thing is, if you already have a AD/DA then the ADA8000 is sort of overkill as the mic-input only routes to the ADAT-outputs, and the ADAT-inputs route to the line-outputs. There is no way to route the mic-ins directly to the line-outs without having an additional DA/AD conversion.
You will have to connect the ADAT-out to the ADAT-in and then from the line-out to the line-in on your A16. Signal would thus have to pass through a AD->DA->AD-stage versus an ideal AD-only stage.
I am thinking about replacing my A16U with two of these but am a bit worried about the conversion quality. The mic-pres are so-so but "transparent" - you could always use a more colourfull preamp before the ADA8000 if desired.
I would rather have a TFPRO 16X, but that is sort of out of my league at the moment. SM audio is cool also, though I really need line-inputs.
Any views on this?
<font size=-1>[ This Message was edited by: voidar on 2004-09-24 17:20 ]</font>
You will have to connect the ADAT-out to the ADAT-in and then from the line-out to the line-in on your A16. Signal would thus have to pass through a AD->DA->AD-stage versus an ideal AD-only stage.
I am thinking about replacing my A16U with two of these but am a bit worried about the conversion quality. The mic-pres are so-so but "transparent" - you could always use a more colourfull preamp before the ADA8000 if desired.
I would rather have a TFPRO 16X, but that is sort of out of my league at the moment
. SM audio is cool also, though I really need line-inputs.Any views on this?
<font size=-1>[ This Message was edited by: voidar on 2004-09-24 17:20 ]</font>
good point. if you have the a16, the ad8000 is a step backwards. the best bang-for-buck mic pre solution is to find an old recoring board with real direct outs. recording board prices are extremely depressed right now. you can find 20 year old $3000 mixers for $200-$600, a bargain for high quality mic pres.......