Topic: Sound Blaster Sound Card good or bad???

[Ed.]

for what its worth about 4 months late...the SB cards are great, just not for transfers.  I love mine, i have the breakout box on the front of the computer and it does a great job and doing what it needs to do soundwise...but i'd never use it for transfers unless i absolutely had to.  but for firewire transfers from the jb3 i'd say its excellent.

[silverbullet]

I may get flamed but for quick jobs when I record DAT 48kHz or something that isn't in 44.1 16bit CD format ,... since SB resamples I can create a 16bit 44.1 master wave file as my masters play. I was doing this for a long time when I was just starting to tape and didn't know any better. I guess the right way is to make a correct master wav file and then have soundforge or wavlab, etc... convert the master wav file to the proper 16bit 44.1 format for cd purposes. 

[caymanreview]

you damn right that would be the correct way

[silverbullet]

While on the topic... I was told for better results to record at a higher rate and then convert to 16/44.1 . Is this true?

If my final product is going to have to be 16/44.1 then I'm currently just recording that way to save steps. But if I can get better results converting 16/48 or 24/96 to 16/44.1 then I might as well start doing that.

[pfife]

While on the topic... I was told for better results to record at a higher rate and then convert to 16/44.1 . Is this true?

If my final product is going to have to be 16/44.1 then I'm currently just recording that way to save steps. But if I can get better results converting 16/48 or 24/96 to 16/44.1 then I might as well start doing that.

Check this out- seems to me the jury is out on your question of running 16/48 and converting- a matter of personal taste and time, I guess.  But the popular consensus is that if you can record at 24 bits, you should definately do that.  I personally don't hear a difference in quality between 16/44.1 and 16/48, but I hear a huge difference between 24/44.1 and 16/44.1.

http://www.taperssection.com/yabbse/index.php?topic=27066.0

hth
~Andrew

[silverbullet]

I reread that thread. I would agree with most that 24bit is much better than 16bit. I need to clarify my question. If I resample 24/96 down to 16/44.1 would the result be better or the same as if I just recorded 16/44.1 in the first place?

[MattD]

Better, unless your resampling and dithering algorithms suck. Whether or not you could hear it once you're back in 16-bit world depends on your ears and playback system.

[silverbullet]

Thanks. It's nice to get a second opinion on what I was told. 

[SparkE!]

Are you planning on using the digital ins for transfers?  If so this isn't the card to go with as it resamples.

I'm going to disagree with you on this one.  I know that the Soundblaster name is not well respected among the taping community, but not all Soundblasters resample.  The Soundblaster Live! series definitely resamples and should be avoided.  However, I have a Soundblaster Audigy 2 Platinum internal card and it does not resample.  I've transferred many, many hours of audio with this card and compared the results with transfers I've made with my M-Audio Audiophile 2496 and they produce identical results.  The way I've verified this is to transfer a recording with the Soundblaster Audigy 2 Platinum internal card and again with the M-Audio Audiophile 2496 card, line them up with each other in a multitrack editing program, invert one of them and mix them down.  The result is absolute silience in the result, with the exeption of minor flaws at the very first of the recording due to the fact that it takes a short period of time for the soundcard to synch to the S/PDIF source.  The Soundblaster Audigy 2 Platinum card usually synchs up more quickly than the Audiophile 2496, so I often get slightly better transfers with the Sounblaster Audigy 2 Platinum internal card.  Once the cards are syndhed to the source, you can't tell the difference between the two.

In my opinion, it's more of an issue of price.  If all you're trying to do is transfer audio from an S/PDIF source, then the Audiophile 2496 works just fine and it has the price advantage.

The reason that I ended up with the Soundblaster Audigy 2 Platinum internal card was that M-Audio was slow to produce working drivers for Windows 2000.  I fought with that company for months on end and finally gave up.  It was almost a year later befoer M-Audio finally came out with drivers that worked properly under Windows 2000.  Regardless, I left the Audiophile 2496 in my Windows 98 machine.  I never put an S/PDIF card in my Windows XP machine because I already had a Nomad Jukebox 3, Hosa ODL276 converter and a firewire card in the computer.  I just run S/PDIF coaxial into the Hosa ODL276, optical out to the NJB3 optical input, then transfer the result by firewire to the computer.  That method also produces bit literal copies, again as compared to transfers directly through the Audiophile 2496 into the Windows 98 machine and transfers directly through the Soundblaster Audiby 2 Platinum internal into the Windows 2000 machine.

Whatever you do, plan to transfer your shows multiple times and compare the results of the multiple transfers to ensure that your setup is making identical copies every time.  It's not always the soundcard's fault.  Sometimes the errors are the fault of the host computer's southbridge chipset, expecially if you are not running the most current drivers and you're using a VIA chipset with an AMD processor.  I spent hundreds of hours trying to troubleshoot my original Windows 98 machine's soundcard only to find out that the majority of my transfer problems were due to a poorly designed motherboard that used an Athlon processor and a VIA chipset.  I finally just trashed that motherboard and started over.  Now I won't use anything but an Intel-based motherboard.  (That's probably like swearing off all Soundblasters, but I guess you learn to hate the companies that cost you the most troubleshooting time and at this point I'm not a fan of either AMD or VIA.)

[mmedley.]

Straight from Tom's Hardware. The whole article is worth a read....

http://www.tomshardware.com/consumer/20021106/audigy2-01.html



The 16-bit/ 44.1 kHz tests gave positive results, though not the same level of quality. This is because when the Audigy 2 resamples all the 44.1 kHz flow rates in 48 kHz, corresponding to the DSP's internal frequency, rates above 48 kHz do not get resampled. So here we have a paradox: the Audigy 2 produces near-perfect sound in 24-bit/ 96 kHz playback (i.e. DVD Audio), but is not as good in 16-bit/ 44.1 kHz playback (such as CD Audio).

You can see the test results are generally positive, except for the intermodulation one. This is evidently due to resampling in 48 kHz.

[SparkE!]

Straight from Tom's Hardware. The whole article is worth a read....

http://www.tomshardware.com/consumer/20021106/audigy2-01.html



The 16-bit/ 44.1 kHz tests gave positive results, though not the same level of quality. This is because when the Audigy 2 resamples all the 44.1 kHz flow rates in 48 kHz, corresponding to the DSP's internal frequency, rates above 48 kHz do not get resampled. So here we have a paradox: the Audigy 2 produces near-perfect sound in 24-bit/ 96 kHz playback (i.e. DVD Audio), but is not as good in 16-bit/ 44.1 kHz playback (such as CD Audio).

You can see the test results are generally positive, except for the intermodulation one. This is evidently due to resampling in 48 kHz.

Yes, it resamples on playback, but not on the Toslink or S/PDIF input when capturing audio from a digital source.  That's where the older Soundblaster Live! series messed up.  The newer Audigy 2 series boards do not resample digital audio as they capture it as a linear PCM (wav) recording.

So, you can make a perfect copy of a 44.1 kHz digital input. What you can't do is play back that perfect copy without the soundcard resampling it to 48 kHz sample rate before it hits the line output of the soundcard.  The wav file remains unmodified.  The 48 kHz resampled version is just used to drive the D/A that feeds the line output of the soundcard.

[silverbullet]

I have an internal SB Audigy Platinum 2 also. I used the digital inputs and had made master 44.1 wav files from 48 DAT recordings as they play. SB had to resample in this case, right? I wasn't recording in 48kHz mode.

[SparkE!]

I have an internal SB Audigy Platinum 2 also. I used the digital inputs and had made master 44.1 wav files from 48 DAT recordings as they play. SB had to resample in this case, right? I wasn't recording in 48kHz mode.

Yes, that was definitely resampled if you ended up with a 44.1 kHz wav file and it doesn't sound "slow".  I'm not sure if it was your soundcard or the recording software that did the resampling, though.  If I'm working with a 48 kHz source, I generally prefer to bring it in as a bit literal transfer at 48 kHz, then resample the resulting file in software.  My current favorite program for doing that is Audacity.  It's free! 

http://audacity.sourceforge.net

Other people swear by Samplitude, but I've never tried that program.  Cool Edit Pro and Soundforge also do a pretty decent job of resampling.

[Ed.]

SparkE!, thats very interesting info.  never knew that about amd's either.

[SparkE!]

SparkE!, thats very interesting info.  never knew that about amd's either.

Actually, I don't think it was the AMD Athlons per se.  I'm pretty sure that the problem was with the VIA southbridge chip and its drivers, but that's what almost all of the Athlon systems used a couple of years ago.  I'm told that they've fixed the problem in more recent chipsets, but I'm not willing to dedicate that type of time again.  Seriously, I spent hundreds of hours trying to troubleshoot that system and it just is not worth it to me to go through that again.  FWIW, the system was built around a KT7A-RAID motherboard with a RAID 1 array of hard drives.  The RAID array failed several times and I think it was due to that VIA southbridge chip.  All of the hard drive traffic, all of the USB traffic and all of the IDE traffic goes through that chip.  I never found any problems with the hard drives, but the RAID array would be broken.  It was a matter of figuring out which hard drive had the correct image on it and cloning that image to the other hard drive.  Nearly every time the array failed, it was during a S/PDIF capture from DAT.  In fact, since I've quit doing audio work on that machine, it hasn't failed again.