SBLive! Driver for Linux

CaNeS was the first of many to write in and tell us that Creative Labs now has a beta driver for the Sound Blaster Live! available on their ftp site. Alas, they are only binaries, but its a start.

Comments

Re:Not a bad start

[Anonymous Coward]

Yes, Creative does own EMU.

Calm down.

[Anonymous Coward]

Creative Labs said they'll probably release the source when it's ready but not while it's a work in progress.

Re:Calm down.

[Anonymous Coward]

The guy who was hiring kernel developers for Creative Labs said something to this effect on the linux-kernel mailing list.

Re:Calm down.

[Anonymous Coward]

The guy who was hiring kernel developers for Creative Labs said something to this effect on the linux-kernel mailing list.

I am that guy, and I said no such thing. Everything is binary-only for now and no decision to release any source has been made.

Jon Taylor
Linux driver engineer, Creative Labs
jtaylor@creaf.con

Re:Calm down.

[Anonymous Coward]

Doh! I misread the post I replied to. My boss made the announcement, not me. But AFAIK there's still no plan in place to release any source. If you want to see source, the best way to encourage CL to alter their policies is to provide feedback on the drivers we release and let CL know loud and clear that Linux support is important for your purchasing decisions, and that open-sourced drivers are likewise important. But if you want to make the most impact, don't say things like "I'm gonna buy another soundcard because there's no source for the SBLive drivers!" That makes people discount you as a paying customer and therefore discount anything you have to say.

Re:Calm down.

[Anonymous Coward]

Huh? Saying that you'll buy from somewhere else makes a company not take your seriously? What planet do you live on in which companies don't care that they're sending customers to the competitors?

Personally, there's NO WAY I'll buy a card that my OS won't support - and that includes SOURCE CODE. Binary-only drivers are worse than no drivers.

A. Nonymous

Re:Calm down.

[Anonymous Coward]

Now that there are Banshee and SB Live drivers ou
you could tell when TNT drivers are out...

Binary only sucks, but its WAY better than nothing.


--
Jsa

Re:Calm down.

[Anonymous Coward]

Binary only drivers are WORSE than nothing. Having "nothing" can't crash your machine but a binary driver can and you won't be able to fix it.

And of course, binary only drivers will only work on a subset of all the hardware out there. (can you say old kernels, new kernels, experimental kernels, non-Intel?) What's more, when a binary-only driver is released, it will work for some people, and those people will have no reason to care about the other poor guys who can't get it working and can't fix it without the source. The manufacturer can then also ignore problems with the driver and simply tell everyone to get "supported" hardware.

In the end, I bet the majority of people running Linux are NOT on i386 machines with 2.0.36pre7 or 2.2.5. In other words, Creative has just released a driver that is NOT going to work on the majority of Linux machines.
I don't exactly call that support.


(however, I don't see anything wrong with them beta testing a binary driver and then releasing the source when it's stable)

Re:Calm down.

[Patrik Nordebo]

When, where and to who did they say this?

Re:Calm down.

[jwhyche]

I may be in the minority here but I really don't get the big deal about a company releasing a binary only driver. I know it goes against the Linux spriit of open source and everything but most people here don't seem to have any problems with game companies releasing binary only games. What's the difference?

Linux's ability to incorp. modules into it is a prefect way for companies such as Creative to support Linux and keep some trade secrets.

Now someone tell me why a 3D card is better than my old SB 16?

Version compatibility?

[Anonymous Coward]

It says that it is only good on kernel 2.0.36, but there is a file for 2.2.5.

Will it work with RedHat 6.0, whose kernel is listed as 2.2.5-15?

Re:Version compatibility?

[Anonymous Coward]

It works fine under RH 6.0 but you have to use the -f option with insmod to make it load. Just follow the directions; I've been playing q3test and mp3s with x11amp for the past couple of hours with no problems.

nick

Re:Version compatibility?

[Anonymous Coward]

Argh! I accidentally deleted the soundcore.o file from my RedHat 6.0 /lib/modules/2.2.5-15/misc/ directory.

Could someone else with RedHat 6.0 please email the soundcore.o file

Jeff
quarters@adelphia.net

Re:Version compatibility?

[Anonymous Coward]

Nick,

I must be thick or something. I can't get it to work. I copied sblive.o-2.2.5 to /lib/modules/2.2.5-15/sblive.o

I modified /etc/conf.modules to read:
alias sound sblive
pre-install sblive insmod -f soundcore
post-remove sblive rmmod soundcore

But, on boot I get a failure in the module dependecies "blah blah blah, sblive.o is compiled for kernel 2.2.5 you are running 2.2.5-15"

Could you (or someone else) give me a clue as to what I've done wrong, please?

Re:Version compatibility?

[Anonymous Coward]

I followed the instructions for the non-redhat kernel install, and I'm getting the same error the other two people were reporting about.. 2.2.5 dependencies versus 2.2.5-15 ... anyone feel like
providing a little illumination on the subject?

Re:Version compatibility?

[Anonymous Coward]

hmm.. well I did both what you said, and what the other guy said, and it boots up now with no errors.. but when I try to play mp3s, it does this freaky half alien breathing, half quake1 ambient cd noise... and when I nuke the window, it continues playing.. its been going for about 10 minutse now, I'm to lazy to reboot. that and the awesome of not being soundless again is just to much to give up after so long :)

Re:Version compatibility?

[Anonymous Coward]

awesome, cd audio now works.. I just can't seem to the mp3s playing, they make really weird noise.. with a deep echo. kinda freaky

Re:Version compatibility?

[Brad Moore]

The driver will work with redhat-6.0 (I'm using it right now). However, the installation script will not. The installation script assumes that the version number of the kernel is 2.2.5 and redhat 6.0's version number is 2.2.5-15.

I attempted to modify the script to use 2.2.5-15 with no luck (I'm not that good at scripts anyways). You're better off following the instructions for non-redhat kernels, and read the LDP kernel pages if you get stuck.

Re:Version compatibility?

[Peter_B]

It's not the soundcore module that needs -f'ing, it's the sblive module. Soundcore should be the right version. I guess that conf.modules lets you specify a 'custom' install method for modules, presumably something like 'install sblive insmod -f sblive' by analogy with the soundcore lines from the README file. So, I /guess/, that way whenever it tries to install the module it always forces it, instead of merely asking it to do it if it happens to feel like it, which it probably won't unless you're using an old kernel.

Re:MIDI... I HAVE SOUND! =)

[Anonymous Coward]

Actually, I like the midi. I don't think it is as good as a GusPnP, but that is just the set of 'SoundFonts' that it comes with. You can replace them to your heart's desire.

I am very happy with this card both in windows, and FINALLY, I HAVE SOUND IN LINUX!!! YES!!!!

BTW: How do I get my SB64 PCI to work?

[Anonymous Coward]

Has somebody got SB64 PCI to work under Linux? I've read all the HOWTO's but they don't say anything about SB64 PCI.

Do I have to configure the card with setpci in order to get it work? And how to proceed after that? The SB support is compiled as a module.

Thanks

Re:BTW: How do I get my SB64 PCI to work?

[bano]

What about a pci128?

Re:BTW: How do I get my SB64 PCI to work?

[tzanger]

I've been trying to get this damn card to produce decent audio forever now.

I've been working with ES1370 and ES1371 cards in 2.2.0-final right through to 2.2.7 and both the built-in kernel drivers AND the OSS drivers produce what I can only best describe as "gargling". this distortion goes away after about 10 seconds of audio but any drive activity starts it back up again.

The motherboard is fairly descent (ASUS P2L97) with BIOS 1005 (1008 is out but I haven't upgraded yet). P2/233 with 64 megs of RAM. The PCI bus is *quiet*... the only other card on it is my TV decoder card which I'm not using at the moment.

Interrupts aren't shared (tried 5,10,11,12), system load is only 0.1 when playing through mpg123 or mp3blaster...

I found an el-cheapo YMF724E-based card and while the output is noisy (low-frequency quiet hard drive noise with no audio playing) the card plays perfectly... ARRRRRRRGH!!!

If *anyone* can help me here, please email me.. I'm going crazy here!!

Re:BTW: How do I get my SB64 PCI to work?

[Quikah]

This is the same card as Ensoniq AudioPCI, so just follow the instructions for that.

Re:BTW: How do I get my SB64 PCI to work?

[Nik Picker]

Yeah I had same problems, I posted a very basic howto on my site. Plus I solved the other integration issues of Ditto and 56k Internals so please have a look
http://www.ashaman.force9.co.uk/nik/linux/linuX- Files.html

Let me know if it helps

timidity. enough said.

[Anonymous Coward]

I have an AWE64 and I still use timidity because I think it sounds better. Just get timidity and all the patches, you won't be disappointed.

mpg123

[Anonymous Coward]

Btw, I found that mpg123 has a horrible jitter with the phantom menace soundtrack mp3's. amp does not have this problem.

debian freaks ruin it for the linux community?

[Anonymous Coward]

Hey, hey, hey. They've said they'll release the source when they are ready. Just search the linux-kernel archives. They have a clearly laid out plan.

Re:Not a bad start

[Anonymous Coward]

Unfortunately, there aren't plenty of good alternatives, assuming you want a 3D sound card. You either buy a SB Live with beta binary drivers or an Aureal based card with no drivers.

beta != work-in-progress

[Anonymous Coward]

Beta implies more work will be done? Hah!

I've seen more drivers and other software die
in beta stage then you can shake a stick at.

I'll save my "WAY TO GO!"'s for companies that have always supported Linux, not those who are
late-comers.

Wrong

[Anonymous Coward]

They've stated the opposite: they have no intention of releasing the source. As such, SBLive users will be stuck with kernels a few versions behind so long as Creative supports the driver, and then they'll be stack with the "last supported version."

no, you're not Jacob Hawley

[Anonymous Coward]

Here is a link:

http://www.uwsg.indiana.edu/hypermail/linux/kern el/9901.1/1398.html

http://www.uwsg.indiana.edu/hypermail/linux/kern el/9901.1/1672.html (refers to SDK)

http://www.uwsg.indiana.edu/hypermail/linux/kern el/9901.1/1675.html (refers to making source code available)

If the searchable database wasn't so crappy, I'd list more links.

Re:Not a bad start

[Anonymous Coward]

Ahh, but 4-Front has released a beta driver for
Aureal Vortex (I, not II). Of course, it's not
open source either, and you do in fact have to actually pay for it. But this is, after all, 4-front's business

So yes, there is an alternative in the 3d-sound world.

No, You're wrong.

[Anonymous Coward]

Search linux-kernel. You'll find they said they'll release specs, an SDK and sample source code.

Binary only...

[Anonymous Coward]

From reading the release notes about the drivers they are going to suck. First of all... No SMP support. Apparently they don't want to take the time to write a good driver. This is probably why they aren't releasing the source. Do they really think that keeping the specs private is going to help their biusness? Oh well... I'm gonna get a 4 channel Digital I/O board where the company is going to release the drivers anyhow.

mirrors?

[Anonymous Coward]

anyone? =]

I didn't think that many people would own an SB live.. but guess enough do.

DOH!

[Anonymous Coward]

I just bought a SB PCI 64 yesterday because there was no live drivers. :~~~~~(
*cry hard*

Re:DOH!

[Peter_B]

Scratch a track off somewhere where it can't easily be noticed, and return the card, saying that it doesn't work. Demand the return of your hard-earned. Buy an SBLive! These drivers work fine. Now I can listen to MP3s and play Quake. Who cares about open-source? They have every right to protect the inner workings of their chip -- and what a nice chip it is. I only wish I had some drivers that would work with NT5B2.

Redhat 6, and KDE

[Anonymous Coward]

I'm running Rh 6 and KDE and i need to know which modules i need to make sure have been removed before loading sblive.o or which ones need to be loaded before sblive.o. if i load v_midi i can get it to load but x11amp core dumps and if i play wave the first one plays but after that it takes 20 seconds for a wave to play and it lags horribly. Thanks.

Posh

Re:Connection Refused - Mirrors?

[Anonymous Coward]

ftp://support.soundblaster.com/pub/creative/beta/

Other mirrors?

[Anonymous Coward]

Anyone snag a copy before that second location went down? Damn I need it bad for Q3Test!

Which card should I get?

[Anonymous Coward]

I want a decent sound card for a Caldera 2.2 box (it might also be running Red Hat in the future).

I will have a single Win95 partition (Ultima Online and Everquest).

I want something that can be EASILY configured with the MINIMUM of time and kernel recompilations. I HAS to work with an MP3 player.

Price is not an issue, I need my MP3's.

Re:Which card should I get?

[Peter_B]

Well, the SBLive obviously works, and plays MP3s. It's also pretty nice in Windows 9x. It's not the cheapest card around, but the Windows drivers have loads of nice stuff (EAX, mainly) that make QuakeII sound so cool it's unreal. Actually, they also make Unreal sound rather cool, too. The card is good quality, and can do 'most anything you'd want.

Re:Mirror

[Anonymous Coward]

Thank you very much.

Re:awesome

[Anonymous Coward]

I'm looking for a 5 channel/4 channel sound card that just support the "left" "center" "right" and "rear" channels. the SBlive is just the studio or 3 channel effect (in real surround sond the rear is just one channel. DVD 5.1 changes that and SBlive isn't 5.1 chanels)..... Other then that that soundcard is useless for professional home systems.

Re:A response to Creative Labs

[Anonymous Coward]

Whats needed is a system whereby developers can
gain source under an NDA, and then (at their
option) after a period of ~12-18 months, make
a source release -- this would prevent the kind
of problems that people like creative a afraid
of (like where Weitek reverse engineered one
of Sun's graphics chipsets because they got
hold of a single .h with the register specs),
since the agreement would dictate what, for the
initial period, could be released, and if you
wanted (or needed) source, you could obtain it
yourself.

supported soundcard? Pass the sauce, Linux wit GPL

[Anonymous Coward]

I read about this sound card on slashdot. I bought
this soundcard and there is pretty good support thanks
to Gerd Rausch. The driver is GPL and available now:

http://www.anime.net/~sam9407

the card is Dream SAM9407 dsp based and available from

http://www.hoontech.com

Too lazy to reboot??

[Anonymous Coward]

Uhh... more likely you are too concerned about your pitiful 'uptime'. Bloody hell, just reboot when you go to take a shit. I assume you don't just sit by your computer and crap into a plastic bag.

Re:Too lazy to reboot??

[Anonymous Coward]

You can talk to my users then when they wonder why they've all been kicked off suddenly.

Re:Too lazy to reboot??

[Anonymous Coward]

Is it too much to ask for more than one person to use a machine? I don't seem to remember any problems using the audio hardware on SPARCstations with people logged in.. why should Linux be any different?

Re:Too lazy to reboot??

[jmalicki]

If you're worried about users you shouldn't even have sound support or a soundcard installed...

Re:Too lazy to reboot??

[Peter_B]

That's wholely irrelevent. The poster was making out that there were hordes of angry users attached to the machine, and that even one lost second of uptime would be a major disaster. If that's the case, then the machine surely shouldn't be used for playing mp3s or Quake. Though I wonder why anyone would bother to recompile the kernel if they aren't going to use it. The person hasn't even tried to use the blinking module -- it works just fine with 2.2.6, though I don't know about 2.2.7 as I can't be bothered to free up enough disk space to let me recompile.

A server in such a situation should be a server and nothing else. It shouldn't be a ghetto-blaster, and it shouldn't be a games console. It should be a server. Just reboot, it doesn't take long. shutdown -r 10, or whatever, but then, a minute or two afterwards, shutdown -r now. Sod the users.

Re:A response to Creative Labs - other platforms

[Anonymous Coward]

More specifially, they WON'T release a driver for non-i386 platforms.

Re:A response to Creative Labs

[Anonymous Coward]

In other words, you won't have a good driver until after the device is obsolete. Good idea.

Re:Mirror

[Anonymous Coward]

Woohoo! You rock.

buying soundcards

[Anonymous Coward]

I love my Soundblaster 64AWE. The only reason I bought it was because it would be compatable with linux. I would love to buy a Soundblaster Live, but until the drivers run on an SMP kernel, I'm sol. I work at a small computer retail store, and I'm amazed at how many people have come to the store looking to buy products on linux compatability alone. Most of these linux users are more than willing to spend more money for a quality piece of hardware. It is really sad that Creative Labs is not able to release the source code, but I hope linux users still buy their products and let Creative Labs feel the effects. I hope they noticed the /. effect they were hit with when they released the drivers.

Opensource is 42

[Anonymous Coward]

Opensource is 42 (for software).

You will learn that Grasshopper.

sblive! problem

[Anonymous Coward]

When i try to run modprob sound (sound=sblive2.2.5
module) i get these errors:

/lib/modules/2.2.5/misc/sblive.o: unresolved symbol unregister_sound_dsp
/lib/modules/2.2.5/misc/sblive.o: unresolved symbol unregister_sound_midi
/lib/modules/2.2.5/misc/sblive.o: unresolved symbol register_sound_dsp
/lib/modules/2.2.5/misc/sblive.o: unresolved symbol register_sound_mixer
/lib/modules/2.2.5/misc/sblive.o: unresolved symbol unregister_sound_mixer
/lib/modules/2.2.5/misc/sblive.o: unresolved symbol unregister_sound_special
/lib/modules/2.2.5/misc/sblive.o: unresolved symbol register_sound_special
/lib/modules/2.2.5/misc/sblive.o: unresolved symbol register_sound_midi

I run debian 2.1 and (of course) kernel 2.2.5.
Please help me, I really want my Live! to start working. What is wrong ?

Re:sblive! problem

[demon]

Did you read the README? You need to have the soundcore module - that has to load before the SB Live driver module does. Read the README.

Re:sblive! problem

[Peter_B]

Well, it looks like it hasn't been installed properly. First, ensure that your kernel is compiled with sound support, but don't make any sound drivers -- just the support. Then copy the pertinent file to /lib/modules/2.2.x/misc/sblive.o . It seems to work fine with 2.2.6. Then make the necessary modifications to /etc/conf.modules -- something like:
alias sound sblive
pre-install sblive insmod soundcore
post-remove sblive rmmod soundcore

because it sounds like the soundcore module isn't being loaded.
And then type something like modprobe sblive. But I'm not sure about that, as I ran insmod soundcore, then insmod -f sblive, because I don't use 2.2.5.

It says all of this in the README file.

Someone help us poor soundless fools!

[Anonymous Coward]

I'm getting the exact same problem
someone help us poor soundless fools!

Re:Someone help us poor soundless fools!

[Non-Newtonian Fluid]

I added this line to conf.modules:

install sblive insmod -f sblive

Though it still reports failure,
it seems to work.

Dream SAM9407 dsp based soundcard

[Anonymous Coward]

the Hoontech SoundTrack 128 DDMA? I wonder if an ALSA driver is coming? Anyone interested in that?

Re:Better yet...

[Anonymous Coward]

This card is not quite as good as an SB-Live, but it is good enough that the sound card will not be the limiting factor for almost anyone (either the speakers or the application will be).

It has 3d sound, with hardware acceleration for more common operations. Part of the 3d is done in software, so it will take up some CPU. This is a mixed blessing, in that 3d audio is at an early enough stage that current algorithms will most likely be outdated in at most a year. The type of stuff the 4dWave accelerates (doppler effects, calculating time and intensity shifts from distance, etc.) will be applicable no matter what the alogrithm. A full end-to-end solution like the SB-Live where you just give (x, y, z) coordinates, velocity, and what sound to play means that a year from now, you'll be stuck with the same old algorithm. The SB-Live has a DSP, so it is probably in theory software-upgradeable; in practice, Creative is more likely to try to sell new cards than upgrade old ones. On the other hand, games with newer QSound drivers will take advantage of newer alogithms on the 4dWave, as well as acceleration for the standard stuff.

IMHO, the only real advantage of the SB-Live is that it has a slightly better codec. Both codecs support better than 44khz 16bit (CD quality) audio, so you really can't take advantage of this yet. However, should anyone ship 48khz 24 bit audio, the SBLive will be marginally better (both codecs are 48khz, but the SBLive one has a few more bits resolution). Really, I'd like to see a 96khz, 24 bit, 6 speaker card at some point.

I'd say the 4dWave is still the best card for a Linux user hands down. The SBLive has a few technical advantages, but they're really minor in comparison to the price, openness and support.

They're not releasing fucking source

[Anonymous Coward]

"SDK, specs for the SDK and sample source" is a bit different than driver source. That means they'll give examples for how to program on top of their drivers, not driver source. SB has confirmed that they would NOT release driver source many times.

You still didn't want the SBLive

[Anonymous Coward]

Return it and get a Trident card. SBLive means you'll be stuck with slightly older kernels so long as Creative still supports the card, and then with the last supported kernel. It's kind of lame that way.

Re:You still didn't want the SBLive

[Peter_B]

They may not support the latest & greatest kernel, but that's not to say it won't work with it. I have no problems with 2.2.6. When they stop supporting the card then there may be a problem -- but at that stage there would be no problem with disclosing the source, because the card would be obsolete. Though I don't really see how a sound-card /can/ become obsolete. It works, it does everything I want and more, it plays mp3s and makes noise in Quake, so what else is important?

Other cards

[Anonymous Coward]

Hoontech also makes a few professional-quality cards that, although aren't designed for 3d audio, should blow the SBLive away in terms of audio quality. You can certainly hook yourself up with a 4 speaker, 96khz 24 bit sound system from them if you're willing to shell out the dough. And they do officially support Linux.

Re:SBlive driver feedback

[Anonymous Coward]

Despite what others have said here, hardware-wise, there's nothing to touch the Live! at this price point, at least from a musician's point of view

Are you serious? I've heard pretty much nothing but bad things about the SBLive!

For example, the card resamples any digital input to 48kHz internally, and you can't turn that off. If you've recorded a DAT at 44.1kHz specifically because you want to burn it to CD, you certainly don't want it to have to be resampled to 48kHz and then back to 44.1. You want a card that will let you just input the 44.1kHz data without any modifications. Or, even better, something like a Zefiro ZA2, which will resample on the fly, but only if you want it to.

Even worse, the card adds distortion to the digital inputs. The point of digital inputs is that you get an exact copy of your DAT or whatever input source you're using. The SBLive, however, will not give you an exact copy (aside from the resampling issue). It will introduce odd spikes at certain frequencies.

Essentially, it may be a good card for games, but for professional use (or even just us amateurs transferring our DATs) it sucks. Spend the extra $100 and get a card that understands the whole point of digital inputs.

If you want more info, there was a web site somewhere that did careful testing of a bunch of soundcards, and you can see the graphs showing the crappy digital inputs. If I find it again, I'll post the address.

Anonymous Coward

Re:MIDI...256 voiced=software

[Anonymous Coward]

192 of the voices on the SB Live! are generated by a software synth. Meaning its not available under linux.

Re:awesome

[Anonymous Coward]

no 3d, and no midi. at least it's a start.

Source release isn't strictly up to Creative

[Anonymous Coward]

The specs for programming the E10K are obtained by Creative, from EMU, under an NDA.

They do not have the power to release the source on their own -- only E-MU can do that.

Re:Source release isn't strictly up to Creative

[Patrik Nordebo]

Hint: Creative own E-MU. If they want specs, they can get them without an NDA.

Re:Source release isn't strictly up to Creative

[demon]

Err? E-Mu Systems was bought, some time ago, by Creative. They did that some time after the AWE32 was released. If Creative wanted to release programming info, they can do it if they want.

Re:Source release isn't strictly up to Creative

[splaytree]

Emu Systems (Emu-Ensoniq as it is called now) is a wholly owned subsidary of Creative Labs. If they want to release the source code to the SBLive! drivers, they can. It's all up to Creative, not Emu.

Re:Not a bad start

[Anonymous Coward]

According to someone I know who works for Creative, they are keeping programming information secret for these cards because many of the differences between the cheap cards and the expensive cards are due to differences in how the DSP on the card is programmed.

The don't want their competitors to easily get see that, and they don't want people to figure out how to make the cheap cards do things that are only supposed to be done on the expensive cards.

Thus, until these are no longer top of the line cards, it's unlikely we'll see open source drivers from Creative.

Good alternative

[Anonymous Coward]

Trident 4dWave-NX based cards are a good alternative. All the features (including 3d audio), but GPL ALSA drivers contributed by Trident.

source code promise here

[Anonymous Coward]

http://www.uwsg.indiana.edu/hypermail/linux/kernel /9901.1/1675.html

other links:
http://www.uwsg.indiana.edu/hypermail/linux/kern el/9901.1/1398.html

http://www.uwsg.indiana.edu/hypermail/linux/kern el/9901.1/1672.html

I assure you, I recall reading more explicit references to making the source code available.

Ordering

[Anonymous Coward]

I ordered mine from Hoontech and haven't had problems (I recieved mine today).

Many smaller companies place orders through a larger company that handles on-line ordering. Hoontech is one of them. It's pretty commonplace, and it's a bit safer ordering through one of the central houses than from a tiny company, actually.

If you're really worried about ordering, I'd wait a few weeks. It's a really new card, so I haven't seen anyone else selling it yet (Hoontech is selling them mostly targetted at OEMs right now). I'd expect to see normal on-line stores selling them soon. It should cost about the same, since the price from Hoontech is $19-$24 plus $14-$16 shipping from Korea. The OEM price is $24, so with a normal markup and shipping within the US, it should come out about the same.

you will learn

[Anonymous Coward]

Binary only are available only for a limited set of processor architectures, get left behind by kernel and application changes and accumulate bugs. These problems are only fixed (if ever), on the schedule of the owner of the source. That is the old world of software that sucks.

Better yet...

[Anonymous Coward]

To all those considering getting an SBLive, I'd strongly recommend going for a card based on the Trident 4dWave-NX chipset (www.hoontech.com makes and sells one). It's a really sweet chipset, and Trident released GPL kernel drivers for it (which will ship with the next version of the Advanced Linux Sound Architecture). It has 4-speaker support, good 3d positional audio, all operations are done internally in 24 bit/48khz, S/PDIF digital output, etc. The Hoontech card has a 20bit/48khz codec, and even optical and coax plugs for digital output for plugging it into high-end recording equipment. Right now, since it was just released, Hoontech has eval pricing, so the card costs $19-$24 (depending on digital plugs) plus $14-$16 s/h from Korea. I'd say it's a much better card regardless of drivers, and having GPL ALSA drivers sure kicks ass.

The problem with binary only (see http://pmitros.mit.edu/binary-only.html during daylight hours) is that Creative is unlikely to still support those drivers 5 or 10 years from now. On the other hand, even source drivers break between releases. Binary drivers are almost guarenteed not to work from release to release (Linus has said so). Your hardware is likely not to work at all with modern software a few years from now, and force you to wait for Creative to update their drivers each time a new kernel comes out today.

- pmitros@SPAM.mit.edu

Re:Better yet...

[Trashman]

Does anyone have this card? How good is the (3d) Audio, and Midi on this card as opposed to the SBLive or Diamond MX300?

Does anyone know of any Reviews?

I may just buy this card if CL or Diamond Don't get on the ball Release the Source.

Re:Better yet...

[Trashman]

Okay, They Openly support Linux,

I'm sold.

I placed the order this afternoon. Now all I need are some good Speakers.

Any suggestions?

I agree, but no source is better than no drivers

[palpatine]

I would definitely rather have an Open Source hardware driver than a binary-only one. But I already own a Dxr2 decoder card (silly me), and if a binary-only driver comes from Creative Labs, I'll be damned if I don't use it. Of course, I'll never buy a product for which the potential for OSS drivers nearly impossible, which makes the G200 a great card to consider.

I have the Matrox MGA-G200 specifications on me, in PDF format. I downloaded them to make a point, and if Matrox changes their mind about the G200, it's too late for them.

Matrox has been making some great moves with opening up specifications and providing the facilities for which Open Source drivers develop and proliferate. Logitech is very much the same, where everything from wheel mice to sophisticated joysticks are well supported by pure Open Source drivers. These companies need to be constantly reminded that they're doing a great job, and they're slowly helping the entire industry do the same.

The point is, Open Source is great, and it's the preferred option. It's not the only option (sometimes it's not even an option, which is the case for the SBLive drivers), but it's by far the recommended option.

Hoontech

[Brett Viren]

Have you found a place on the web to order this card which doesn't shunt you through half a dozen domains (as Hoontech's site does)? I'm okay with sending c.c. info over the web, but not when the path is so tenuous.

-Brett.

Re:MIDI...

[Nick Ives]

The MIDI support on this card is at least as good if not better than the AWE32/64 IMO. My sister uses it for all of her home studio setup under 'doze and it works very nicely along with Cakewalk to give a nice MIDI setup.
I have on in this box aswell, I used to have an ISA card in here for sound in Linux, but I just ripped that out as this driver for the Live gives me all the audio capabilities I need for Linux (wave and CD Audio). Now, all I need is EAX support with this driver (I presume it doesnt have it....) and a few more games. Then I can get rid of that pesky Windoze partiton....

Nick

Cool

[drwiii]

Maybe I'll look into getting one. For the time being, however, my 3 year old SB16 PnP is working perfectly.

P.S.- Canes: :P~

Re:Cool

[CaNeS]

:p~~~ drw

the drivers are pretty smooth too ;)
i've been playin q3test with em since yesterday ;)

Re:Cool

[_Lewellyn]

IMHO, for the time being the SB16 is probably the best choice for most Linux users. (No flames, please!)
I haven't found any other card that works quite so easily and reliably, let alone how cheaply they can be bought.

Re:Not a bad start

[J4]

Not EMU... they bought Ensoniq

Re:Don't trust 'em!

[Peter Amstutz]

As far as I know, they specifically hired a fairly experieced Linux coder. This isn't just getting a few of their internal engineers to figure out "this Linux thing", this is someone who knows what he is doing and pushing hard from the inside to get Creative to support ALL of their hardware on Linux. So I don't think the situation is as bad as you think.

Re:Close, but no cigar.

[knghtbrd]

Good, do that. But by the gods, keep your letters POLITE!

You want them to know you chose hardware with open specs and drivers with source sure, but you don't want to go out of your way to piss them off!

A response to Creative Labs

[knghtbrd]

If all of you were expecting a big, long letter to Creative talking to them about the advantages of Free Software and Open Source, you won't find it today. Oh I could spend the time to write all of that, but only their techs would likely understand it fully and their suits would just not get it. So I'll skip the details on why and just focus on the what:

Creative Labs,

I am pleased to see that you are taking an interest in the Linux platform. I am also pleased that you are now releasing driver support for one of your most interesting products, the SB Live. However, I regret to say that I have not purchased and do not plan to purchase your product at any point in the near future because of what I believe is a lack of sufficcient support for your products. I have seen and was impressed by the SB Live, and would happily reconsider my decision not to purchase your product should the issues I see with support be resolved.

The number one issue I see with support is source code. Without it, I could never right mindedly purchase your product. The only way to use the SB Live is with a software driver. On the Linux platform, these drivers are in the form of kernel modules. Without source, I must wait for you to release a new module every time I wish to upgrade my kernel. This delay is unacceptable to me, and I will not limit myself to the kernel versions you have decided to support.

Additionally, binary only kernel modules pose a significant security and system stability risk. I am not saying necessarily that bugs would not happen--they will. And I'm not saying the coders in the community can fix a problem once discovered any quicker than your in-house coders could--likely they can't. What I am saying is that with published code there are more eyes watching. Bugs are found quicker simply because more people can work to track them down.

Please consider it. The Linux community tends to best support those who best support us.

Close, but no cigar.

[HoserHead]

I've said it once and I'll say it again: Binary-only is not good enough.

I've got a plan for the next computer I build. It will have a Matrox G200, because a) it's probably the best 2D card out there, and b) Matrox released all the specs for the G200. I will be sending a snail mail letter to Matrox & ATI, and possibly Nvidia (depending on their TNT 3D path - source or no source) telling them exactly why I bought what I did. The same thing goes for the sound card. I'll be buying one based on the Trident 4D wave chip - because Trident themselves developed an ALSA driver for it. Creative and Trident wil be getting letters on that.

Why am I planning to go to such lengths? Because I want people to know I support companies who support us . I might never modify my GLX module (or whatever) for the Matrox G200. But I want the ability to do that. I want to know that if I buy myself an Alpha, that it will probably work on that too - and if it doesn't, I can probably make it work there.

Creative: If you want Linux people to buy your cards, give the source of this to Alan Cox for inclusion in his -ac series. It will rapidly get tested, bugfixed, and then I will consider taking you off my 'blacklist' - where you currently enjoy the company of corporations such as ATI. I influence the purchasing habits of a large group of people, and I consistently tell them not to buy ATI cards - because better cards abound. I reccomend against Creative, too.

You can have your intellectual property - but you can't have my money.

Re:Close, but no cigar.

[Lee Cremeans]

I've got a plan for the next computer I build. It will have a Matrox G200, because a) it's probably the best 2D card out there, and b) Matrox released all the specs for the G200.

I got a G200 based on this same idea -- XFree86 supports it beautifully in 2D, and there's already beta GLX driver source for it around (though it's a pain to get working, and I haven't taken the time to do all the necessary compiling). Quite an improvement from the 2MB ATI Rage I had in here before (I can do 24-bit sparse color at 1152x864x67Hz now, yay!).

-lee...I really wanna see q3test at a speed faster than 5 seconds-2 minutes per frame without having to buy or borrow a Voodoo...

Binary-only drivers

[cduffy]

The problem with binary-only drivers, like binary-only applications which must run suid root, is that they can potentially crash your machine -- and you can't do anything about it, being unable to properly debug and fix the problem.

That's the argument, at least. I'd still call binary-only drivers better than none at all if they're for hardware that's not likely be installed on a server (the SB Live! being an excellent example).

As far as a 3D card being better... it accelerates lots of stuff that would otherwise have to be done by the CPU regarding directional sound.

Modules would have to be converted before load

[cduffy]

Emulating kernel modules at runtime Just Won't Work.

Now, it might be possible to write an emulator that would save a translated binary (okay, it's not quite an emulator per se, though it would share a _lot_ of code)... O'course, the resulting binary would be liable to be quite inefficient... though (now, I don't know alpha assembly, so take this with a grain of salt) I'd expect it'd be quite a lot faster than emulating Alpha apps on an x86 (since the Alpha has more registers... no?).

Does any software like what I just described exist?

Re:Modules would have to be converted before load

[krynos]

The Alpha is able to emulate a x86 to initialize PC hardware in your machine (SCSI cards, Video cards) when the system is booting and in MILO... The Linux kernel itself doesn't emulate the x86, altough there is em86 for user mode binaries.

The Alpha as 32 64bit registers, while the x86 as about 6 usable registers, not to talk about 64bit adress space. I wouldn't want to emulate an Alpha on a x86 machine, the performance would be abismal and emulating 64bit registers and adress space would be a major pain.

Re:Modules would have to be converted before load

[GargoyleMT]

Just Won't Work? True enough. However, it is not beyond the realm of possibility, just merely very difficult.

Or maybe it is impossible, and I just don't know what I'm taking about.

This is the problem when you try to answer lusers's questions, and not dumb the answers down.

Alpha kernel space x86 emulation (sort of)

[cduffy]

Could em86 be modified to save converted binaries rather than doing so on the fly? (Can/does it do that now? I really don't know). If so, could one not convert kernel modules in a similar fashon?

Just writing a program to read disassembled x86 source and output alpha assembly seems to me that it wouldn't be that difficult. Such a program could then be used to convert binary-only kernel modules, no? (Though I suppose that in drivers there might be architecture-specific constants and the like... if enough arch-specific stuff is done through calls to the rest of the OS, though, it might still work)...

Ahh, nothing like a bit of uninformed rambling to start my day. :)

Grrrrrrrrr!

[volkris]

I want my MX300 drivers!

It was a tough choice deciding between the SBLive! and the MX300, and looks like I picked the wrong one.

OMYGOD! IT'S NOT OPEN SOURCE! RUN FOR THE HILLS!

[jonr]

Sheesh, some of you Open source advocates need to pull your panths out of your asses and relax a little! :) The world is not coming to an end. CL gives the Linux communti a small helping hand, and you people bite it! I really don't care if my drivers are open source or not. If they work, fine, if they don't, I wont use the hardware they are written for, it's that simple. *shrug*.

Open Source is not 42.

Jón

Re:Well, at least this is something.

[ptomblin]

It works all right for me. Catting a file directly to /dev/audio just produces horrible noise, but using the "play" program works fine.

I tried to install real player 5.0, but it gets a segmentation fault while playing welcome.rm.

I also tried playing xboing with -sound, and the sound lagged severely behind the game. But I've never done that with a different audio card, so I don't know if that's a problem with the card driver or the game.

Re:SBlive driver feedback

[Tarrant]

I've definitely been having problems in 2.2.7. Sometimes the module works, and sometimes it freezes anything that tries to write to /dev/audio. I really wish I could just compile the damn thing.

Even if it's not GPL'd, it would have helped a lot if Creative released source.

Re:Not a bad start

[Q*bert]

I agree, and I won't be buying one of these until they do release the source. Folks, let's our money where our mouths are: If there isn't an open-source driver available, don't buy the hardware. There are plenty of good alternatives to this thing (like the Yamaha OPL3-SA2 on my motherboard)!
Beer recipe: free! #Source
Cold pints: $2 #Product

Quake3 test && mpg123 && Redhat Ke

[jelwell]

Well, the sblive driver worked great for Quake 3 test; but, for mpg123 all I heard was crackling and chirping. At least the important things worked.
By the way, the redhat kernel source rpms are no good because the kernel version needs to be 2.2.5 not 2.2.5-14 or 2.2.5-anything. So go grab the kernel sources at www.linuxhq.com and uninstall the kernel-source rpm.
Joseph Elwell.

Re:Mirror

[wo0zy]

Thanks a lot, it is much appreciated

Connection Refused - Mirrors?

[Ezra]

Geez, theres gotta be a better way to do this sort
of thing ;)

We /. everything!


Anyone have a mirror or care to post one?

Thanks

Re:Finally

[Ezra]

echo "Soundblaster Live! driver installed. Please reboot."

First of all, i didn't even see that "Please reboot" part, and second of all, i'd just like to say that's the easiest it's ever been for me to get a piece of hardware to work with linux. I think it was very thoughtful of them to provide an install script...
And, if you would look at the code (below) you would see that their were two possible types of install, one which needed a reboot and one that did not. Why don't you thank them for makeing it so easy for you to install it? (yes.. im beggining to forget the months and months i've waited for them to release this BINARY-BETA driver..)

Heres the install code:
if [ "$OLDSND" -o "$OLDMID" ]
then
echo "Soundblaster Live! driver installed. Please reboot."
else
if modprobe sblive
then
echo "Soundblaster Live! driver installed."
fi
fi

do insmod -f + if we had source everyone canuseit

[Drizzt]

just force the driver insertion and it works fine!

And btw,

if we had the source code not just people running two kernel versions and non-smp could get this to work. In any case Creative, you are 25% of the way there, just go the extra mile.

Alan
http://www.denizengames.com

Re:OMYGOD! IT'S NOT OPEN SOURCE! RUN FOR THE HILLS

[Rick_T]

| I really don't care if my drivers are open
| source or not. If they work, fine, if they
| don't, I wont use the hardware they are written
| for, it's that simple.

And that's the problem with binary-only drivers - especially those that may be tied to a particular kernel release. If I upgrade, my card might not work anymore. The SB Live isn't a cheap card. Before I lay out the price Creative wants for one, I'd like some peace of mind that when I upgrade my kernel from 2.2.5 to 2.2.7 or beyond that it'll still work. With the state of the drivers the way they are now, it's too much of a risk to take to purchase one, as I don't use Windows at home at all - and I don't want to worry whether my card will suddenly turn into a paperweight with a simple kernel upgrade.

Oh, if you already *have* a SB Live, this driver is potentially a good thing. But my hardware decisions are based on their being good Linux support available. As of yet, that's not there for the SB Live. It's a start, but if Creative wants (more of - my current soundcard is a Soundblaster 16 from 1994 or thereabouts :) ) my money, there's got to be a better driver.

Just how is it supposed to work?

[Cactus]

I just can't figure out how Creative plans to stay current with the kernel. I mean they'll have to distribute the binary verison for all the architectures times all the kernel versions (including -acX and -preX versions) times 2 (SMP and single CPU). Will they provide that much flavours?

For example I can't try this because I'm running 2.2.6 (I have 2.2.7 compiled but am just lazy to reboot:) on an SMP box. Any ideas what I should do?

Re:Just how is it supposed to work?

[Stardate]

Follow the manual installation procedure in the README file, but if you want to use it you'll have to load it manually with 'insmod -f sblive', I have it running right now under 2.2.7 and it's fine. I'm just going to put that in my rc.local... I'M SO HAPPY X11AMP WORKS!!!!

Not a bad start

[Booker]

Not a bad start, but I sure wish they'd release source on these things. Precision Insight has noted that

Experience has shown us that once a hardware manufacturer sees the positive response they receive from the Linux community when support for their hardware is available, they will ultimately agree to allow the release of the driver source code.

It'd be great if CL moved in this direction.

Re:Not a bad start

[Booker]

Ah, hadn't thought about that... it gets even more complicated when there are more fingers in the pie. :-) But then... didn't Creative actually buy EMU?

Re:Not a bad start

[MikeTurk]

But isn't Creative under an agreement with EMU (their chipmaker) about their chips? Are they legally able to release the source, or even the specs?

Mike
--

Re:Don't trust 'em!

[Black Rose]

They also spat out beta drivers for AWE64 on OS/2 and then sat on them forever without releasing anything. In my experience, they just waffle around and do whatever appears to be popular at the moment. If there's a bug, they never seem to keen on fixing it. Seem to always blame something else. If I recall correctly, there was also a minor little war going back and forth when they were trying to blame IBM for lack of SB drivers somehow... They have no concept of loyalty to customers, unlike other companies I know, which are much more deserving of my support!

Did any company, even IBM, support OS/2 as much as the people that used it would have liked?
I don't think it's fair to single out CL about the lack of support issue, Microsoft's just as bad. 8)

what a...

[Signal 11]

Sheesh. After we have already incorporated all the drivers into the kernel, and have several companies (www.4front-tech.com) supporting the soundblaster... they just /now/ decide to release a driver.. binary only, no less?

--

Re:what a...

[{X-Frog}]

SB Live! is not like other SB16 and AWE, not the same architecture (SB Live is PCI...). SB Live! had no drivers before, except the one of 4front technology that was not in pre-alpha release, it was just playing the CD music...
Way to go to Creative Labs!

Creative Labs

[Accipiter]

They're finally moving in the right direction. This is good, to see a leader in the industry hit the Linux community like this. It may kickstart a domino effect to force other companies to do the same. Let's hope so.

WAY TO GO CREATIVE!

-- Give him Head? Be a Beacon?

I agree

[GargoyleMT]

I contacted Diamond for drivers to mine, and they told me to go to http://www.suse.com/. Not that they're of any use, for sound drivers for the MX300.

Aureal is next on the list - I want ALSA drivers for the Vortex!

Any other MX300 owners interested in voicing their opinions to Aureal/Diamond?

Re:I agree

[Shad99]

sure I want a MX300 driver one more reason to remove my AWE64 that I only use with linux...
I've emailed both but haven't heard anything...
I choose the MX300 for the wide variety of formats it had the potential to use (A3D 1.0 & 2.0, DirectSound3D, & could support EAX). Pity they aren't terribly interested in supporting linux as well...

Re:ALPHA and other platforms

[GargoyleMT]

This isn't source that you can just re-compile. You can probably arrange some sort of emulation to get them to load, but I imagine that is complicated by the fact that these are modules, so the emulator might have to be a bit non-conventional.

In theory: Yes
In practice: No

Of course, I don't have an Alpha running Linux, so this is all (as is typical on Slashdot) speculation.

Finally

[Rui]

It's 3:30 AM and I finally got it. I figure 60K after 8 hours of trying gives an outstanding transfer rate of 0.0021 Kb/s :-)

On another topic, I'm not totally against binary only drivers but installation scripts that go like:

echo "Soundblaster Live! driver installed. Please reboot."

give me the creeps...

Anyway, I can't complain. I had the board now I have the drivers. I am a happy man!

It's just more "Shut up" software

[Geinus Roy]

Like WP8. You can't use the copy / paste features because they think Linux doesn't have one, even though X does and the Linux part isn't really important. Not to mention they didn't even bother to get a current version of Linux to develop on. Then Grand Master Linus says, Good Job Corel. That's what they wanted. For us to think they give a rats ass. Same with CL. They'll release them and make a whole lot of Linux users happy and they'll have our support.

Re:It's just more "Shut up" software

[Geinus Roy]

Forgive me if I don't trust them. Yes, it could be
that they will comply completely and they'll get the drivers to work great and all. That would be lovely. When I see a company supporting free software, I get a bit curious. When everything is final and works well, and they release the source and it's in the kernel, _then_ I'll say good job CL. Until then, I just don't know.

Re:It's just more "Shut up" software

[Drayke]

Trouble is that when the driver is buggy, version-specific (herein lies its biggest problem and the biggest issue with being binary-only), -and- they're holding their cards close to their chest, well, they're not doing themselves any favours either 'cos it means they're doing Linux development half in the dark (99% of their development team is still Windows-oriented and will remain that way). By making the thing binary-only, they're screwing themselves out of the best development team they could have, namely the people who, whether for money or whatever, deal with and develop for Linux on a daily basis, and therefore know the APIs they're developing for and so on.

Personally, I'm inclined to stay with the consensus on this one, that it's not enough. Granted, Creative's worried about their proprietary information and so on, but why have they been dominant up to this point? Not because they alone implement the standard, but because they provide the best, most compatible implementation. Opening it up gets the developer base, gives Linux users a good reason to buy an SBLive, and while it opens up the interface to competitors, it also means that people are going to take the time to make that driver something worth using.

Re:It's just more "Shut up" software

[Scott_F]

sorry, but this point isn't taken at all. how can you determine whether they are sincere or not? first off, at least they released it under "beta", which implies more work will be done. if they released it under "final" and there are the problems that i've read here, then you can talk all you want.

it is this kind of negative "rebellious" attitude that can bring down the support for the linux community. notice the keyword there though: community. they support us because we want it, and for them to take up the reins and write even just a beta driver is commendable.

sure the source code would be great, but the "suits" are having a hard time with being rational. :)

i say way to go creative. it's not taking a risk; it's taking a step towards the future.

-=Scott

Re:It's just more "Shut up" software

[razorwire]

Maybe it's just me, maybe it's just that I'm new to the Open Source community, but I don't understand this kind of reaction. It seems to me that the Linux community should welcome the efforts of a major hardware manufacturer like Creative, instead of saying "Screw you, it's not free enough".

Creative seems to be testing the waters here, rather than jumping into the deep end; just because we know Linux will achieve World Domination[tm] doesn't mean the suits buy it (yet), so they're perhaps not so eager to pour huge amounts of money into a software product that, in their eyes, isn't a huge potential moneymaker, and will be given away, source code and all, to boot!

Believe it or not, not everyone is convinced yet that Open Source is a Good Thing, and you're certainly not going to change anyone's mind by telling them to go to hell...

Re:Doesn't matter

[dirty]

Wait, so you make a contract w/ company X to not release the source to something. Then you buy company X. You still cannot release the source for the product? Who is going to sue you? Company X? But they are now you, so why would you sue yourself.

The thing is, given the american legal system, i can see this as being 100% true.

MP3s?

[Robotech_Master]

So I got the drivers installed--and praise the lord, sound on my computer in Linux for the first time since I put it together.

However, I'm having problems playing mp3s. mpg123 does the following:

opening socket, format = 0x00001021 at -1 Hz
unsupported playback rate: -1
audio: Connection refused

x11amp segfaults, even after recompiling from srpm, and freeamp seems to work all right, but it won't take m3u file inputs.

If someone could email me with suggestions, I'd greatly appreciate it.

Re:Well, at least this is something.

[Quikah]

If you are using gnome try disabling esound. I had this problem you describe when using esound. Without it works fine.

Re:Well, at least this is something.

[Quikah]

Kill esd process for a quick fix.

To turn it off in gnome go into the control panel and uncheck the sound support. I am not sure, but if your running E you might need to disable sound in that also, open e-conf and uncheck the audio option.

Mirror

[Quikah]

People seem to be asking for this, so I threw it on my ISP account webspace, I don't use it anyway, so what the heck:

http://www.ecom.net/~jimz/sblive-0.1b.tar.gz

Re:Mirror

[Real Scumbag]

Champion....

drivers work pretty good

[oriel]

i have an SBLive in my box right now and the new drivers aren't that bad. i mean the install went smoothly and i'm listening to x11amp right now. RH5.2, KDE, kernel 2.0.36

but i dont hear too big of a difference between the old SB16 i had in there yesterday, so oh well. hopefully we'll get some development and source for these things.

Re:awesome

[Th0th]

Actually, the live supports 4 channels, not three. It has front left and right, and back left and right. Dolby Pro-Logic (but not 5.1 surround)

MIDI...

[fordp]

I've been looking at this card for awhile, because it looks like a nice card. I've recently been told that the MIDI support on the card is substandard. (as in not as good as either the AWE32, or AWE64) Can anyone verify this?

The last time I spent more than $50 on a soundcard it was a SBpro, and I really don't want to repeat that mistake. :)

Re:MIDI...

[eelke]

I think the MIDI sounds exactly the same as the AWE64/32 if the soundfonts are the same (my awe64 had only 512k of ram, so my live is better).
It isn't possible to enable any effects for a whole song though (per channel in sequencer)

Re:MIDI...

[ArchAngelQ]

Actually, it has better midi, as far as it matters. I has the same hardware and system as AWE 32 and 64, but has support for up to 256 simultanious voices. That's the only diff, as far as I know, and 256 is far more than you need, of cource. You would need that if you where going to comploses a symphony in midi.

awesome

[mTor]

I always wanted to buy that card but the lack of a driver stopped me from it... Anyone know if the driver supports 3D sound features?

mtor

Re:awesome

[Gr00ve]

Playing mp3s, cds and quake is all I want mine for.
Midi? Give me a break.
Are there any games that use 3D sound out on Linux?

Right, I'm off to tell Diamond (MX300) why they lost another customer...

Got a working PCI64/128

[rbb]

Getting your PCI64 to work is actually quite easy.

After messing around in the kernel for quite some time with the Ensoniq drivers it didn't quite work the way I wanted it.

So i thought "what the heck" and tried the OSS package from Open Sound. Works just the way it should.

Okay okay, it's payware but hell the rest of my Linux is free anyway so I didn't really mind the concession.

If you can spare the $30,- (20 for the core and 10 for the PCI64/128 support) than it's WELL worth the money.

-rbb

Re:no-DOH!

[Chad Page]

There, there... it's not so bad, really. :) It's not a bad card at all really, once you have the PCI concurrency issues worked out (for me, 2.2.x>3 or so fixed it, YMMV) it works nicely and installs easily (esp. compared to the Dark Ages of ISA boards, esp PnP.)

Doesn't matter

[Hamshrew]

Even if Creative bought EMU, the contract is still valid. If it says they can't release the source, then they can't until the agreement expires. I suppose it would be possible to get the agreement nullified, but the amount of paperwork involved is probably staggering.

Re:Well, at least this is something.

[datawrangler]

OK, I'll bite. How do I disable esound? I've tried running x11amp and mpg123, but the MP3s play both slowed-down and stuttery.

Well, at least this is something.

[PhoboS]

This is a step in the right direction by Creative, but the driver still needs a lot of work.

As the README states, it currently only supports kernels 2.0.36-0.7 and 2.2.5. I tested the driver on my system with kernel 2.2.5, but the result wasn't very useable. The install worked fine, but the sound was only unrecognizable noise, that also seemed to be lagged a few seconds.

I hope that they work out the problems quickly, and that they will support all the features they have under Windows.

Re:Well, at least this is something.

[PhoboS]

I guess I should have done some more testing, I only tried playing an mp3 using mpg123, and since that didn't work I reverted back to my old sb-16.

I'll have to try more stuff later, but since I use mpg123 alot I really hope they will fix the bugs soon.

No source, no purchase

[chris.bitmead]

I'm in the market for a sound card right now. If SBLive had source drivers I would be getting it because it's a great card. But no source, no purchase, so I'll be getting something else.

Don't trust 'em!

[Atomic Frog]

They also spat out beta drivers for AWE64 on OS/2 and then sat on them forever without releasing anything.
In my experience, they just waffle around and do whatever appears to be popular at the moment. If there's a bug, they never seem to keen on fixing it. Seem to always blame something else.
If I recall correctly, there was also a minor little war going back and forth when they were trying to blame IBM for lack of SB drivers somehow...
They have no concept of loyalty to customers, unlike other companies I know, which are much more deserving of my support!

Only in the best case, if Linux does take over the world, then Creative will have _no choice_ but to support it. But otherwise, watch out!

And no SMP, either!

[Ventifus]

The drivers don't support smp kernels. Well, I'll give it a shot anyway, though I don't expect any luck. I wonder if there's any way of wrapping a module so a uniproc compiled module can be used with a smp kernel?

Other Kernel Versions

[bbarrett]

Has anyone had any luck with running the module on other kernel versions? I currently am using it under the 2.2.6-ac2 kernel.

To install, i just ran "insmod soundcore" then "insmod -f sblive" to force the install. Bad things will probably happen if any sound modules are already loaded, so watch out for that... Esound doesn't work right, but I can use x11amp by sending the data through its OSS Plugin (the default anyway). Esound just garbles everything more than a little boing. Of course, I have seen a couple reports here that it is doing that on a 2.2.5 kernel anyway....

Guess I'll have to take out the SB16 that I have had installed for Linux compatibility... woo hoo.

I understand why it doesn't work...

[ArchAngelQ]

Or, at least I think I do. I just got my card yesterday (5/7/99), and installed it under windoze first, so I know what it sounds like with the full drivers. I had heard on here that it could be a bit of a pain to install under linux, but this was not entirely true. I installed it with the script, and it went fine. Granted, I'd read the readme, and compiled my latest kernel with the support it needed, but I figured, why rebuild it a second time when I need to add printer support anyway. However, the first mp3 I played in x11amp came out choppy, and I thought I was hosed. However, I tried another one (higher bandwidth, and standard one too), and it came out fine. Or, rather, it came out like I'd played it in a SB 16. Which is when it hit me. The people at Creative don't have an experienced Linux person (see there job offer for a linux porter, elsewhere on slashdot), and as such, did there best with what they had. So they implimented the SB16 emulation under linux. Sure, it doesn't meet peoples expectations, but, as thier job offer shows, they are serious about makeing it work. Good for them. I know people had feared that they might do what they did with OS/2 support, but, I think not. Just my two (three... four... five) cents, hope it helps some understand why it sounds crappy a bit. Oh, and it is hardly, or just not, optimized, so it plays poorly with other cpu intensive apps, like snes9x.

ALPHA and other platforms

[nerijus]

I downloaded a driver and cannot see any word
about supported platforms in README. Will it
work on Alpha?

Re:A response to Creative Labs - other platforms

[nerijus]

You definitely should note that with binary only drivers manufacturer has to provide drivers for a lot of platforms, not only for i386.
There is no such problem with open source drivers.

I'm torn between awesome and sucky...

[graspee]

Well, first of all I am overjoyed that I can listen to
LONNIE LISTON SMITH while I am coding. Hearing sound
at all in Linux is a novelty for this lemur in 2.2.x kernels.
(When I ran 2.0.x kernels I had a SB32 and had to initialize
it in DOS then loadlin in...)

The big sucky feature is the lack of features. Basically all
this module does is emulate a SB16 very badly. I personally
have got CDs to play, and samples with l..o..t..s.. of
dropouts, like when you have your DMA settings all wrong
in Windows. There is apparently no support for using all
4 channels on the card, or using environmental audio,
or loading sample banks for MIDI.

I am sure that someone will discover how to stop the
samples doing weird stuff though, since it seems to
work for some people.

I really don't see why CL won't release the source though.
I mean, the more people have access to the features of
the sound card, the more its popularity will grow as more
operating systems and applications support it. Weird. Is it
supposed to cut down on evil rival soundcard makers
stealing industrial secrets or something? You can just see
them cackling with glee over the sourcecode:

"Aha! So they *do* use Port !# to access the reverb
amount! Hahahahahah! Quick, make ours do the same,
this is obviously a GREAT port to use..."

Slashdotted?

[wct]

I keep getting a "Connection refused - the server may be down" etc. message from Netscape. Could someone who downloaded it put it up somewhere? I've been hanging out for this for so long - now I can finally test my project (a virtual drum machine) on my better sound card...

SBlive driver feedback

[wct]

Well I downloaded the driver yesterday and played around with it a little. On my RedHat 5.2 system under the 2.0.36 kernel (I've still got it lying around on my system) it worked OK. It's still quite choppy around the edges, and compatibility isn't great - IMHO calling it a beta may be a bit optimistic. x11amp works fine, but many other apps, including esound, mpg123 and my own project, which uses the STK streaming wave class, have problems ranging from degenerating sound output over time, to severe choppiness. On my recently

As for the binary driver issue, to be honest, before I downloaded it I couldn't care less, as long as it worked. Despite what others have said here, hardware-wise, there's nothing to touch the Live! at this price point, at least from a musician's point of view. Unfortunately, despite my best efforts (involving lots of /etc/conf.modules tweaking), I wasn't able to get the 2.2.5 driver (the latest one included) to work with my just-compiled 2.2.7 kernel that was prepared specifically to work with it (no SMP support, no installed sound drivers but sound driver support).

Sure, it's only a beta. But if they can't get a 2.2.5 driver to work on a 2.2.7 kernel, this doesn't bode well for the future - personally, I don't like the idea of having to hold on to an insecure kernel just so my soundcard works. There were 36 patches of the 2.0.x series...you do the maths ;)

In the end, though, I'll reserve judgement for the finished product. At least I have clean MP3 playback under Linux now...one step closer to trashing that Windows partition*sigh*

Re:SBlive driver feedback

[Peter_B]

Isn't it just a matter of insmod soundcore, insmod -f sblive? After putting the file in /lib/modules/2.2.x/misc, and renaming it to sblive.o . That's what works with 2.2.6, and it's what other people have said works with 2.2.7.