VOGONS


The "12-bit" Sound Blaster 16 Myth

Topic actions

Reply 120 of 129, by 640K!enough

User metadata
Rank Oldbie
Rank
Oldbie
SirNickity wrote:

AdLib and Gravis created better products. (Let's just assume that's a fact rather than subjective.) They cost less than contemporary Sound Blasters at the time. They are no longer with us. Think about it -- why do companies usually fail? They can't reach profitability (selling for too little, or their cost is too high?); they don't have sufficient market penetration (they aren't marketing enough, or at least effectively enough?); their product is inferior (but we ruled that out)...

IMO, AdLib failed because they weren't aggressive enough at marketing (not just ads -- which they did -- but developer, retail, and customer outreach for e.g.) They also kinda stumbled into the market when there wasn't any competition to speak of, secured their place, and sat there for too long before advancing. So they got bowled over by someone moving faster. "But but the Yamaha thing!" Yeah, but why did Yamaha agree to those terms? Because Creative had more sales potential, so they went where the money was going to be.

For someone who claims not to be defending Creative, you're certainly acting like a devotee to the Church of Sim. Are you really trying to dismiss the historical record and pretend that Creative survived because they were the only company that understood how to price their products? If that's the case, I don't think there are any facts that anyone could present that would convince you otherwise.

If you think about it objectively, Creative was never a proactive company in the sound world; they were purely reactionary, preferring to sell the same over-priced products as long as they could, until it became clear that they had to do something if they didn't want to become irrelevant. Then, they would try to get rid of more competitors, throw together a half-baked response, and wind up the marketing machine and arm-twisting to get it supported.

From the information that was available, Ad Lib was working on the Gold as a response to the Sound Blaster. If not for the purported Creative shenanigans, it was scheduled to ship around the time of the dual-OPL2 Sound Blaster Pro, if not earlier. Developers at Ad Lib were the ones who approached Yamaha and helped define what would become the OPL3, so it would make sense that theirs was just about the very first card to include it. Supporting that is the fact that Dune shipped in mid-1992, with a 4-operator, stereo soundtrack that was exclusively supported on the Ad Lib Gold, effects module and all. This also shows that they were working with developers to gain support.

As I think history will show, Yamaha trusting Creative in their arm-twisting efforts was a big mistake. It may have helped in the short term, with Creative buying many chips, but where did it get them? Creative subsequently got E-mu to develop CQM, and with few other major sound card-makers left, Yahama's influence on PC audio waned. At this point, we can only speculate about how PC audio might have evolved had Yamaha not given in to Creative's demands. What could Creative really have done, had Yamaha told them to get lost? Having a card without Yamaha FM at that time would likely have had the same market impact as the Game Blaster (dud), unless you could deliver MT-32- or Sound Canvas-like quality at nearly FM prices.

Reply 121 of 129, by Scali

User metadata
Rank l33t
Rank
l33t

If you ask me, AdLib failed because the AdLib Gold was too ambitious.
I think if AdLib had a generic SB-clone in-between, a delay on the AdLib Gold would not have hurt them that much.
But if I recall correctly, the real problem is that the AdLib company was not a company of technical people, but rather musicians/people with a vision. The AdLib card was designed for them under contract. So they could not design their own hardware in-house. Which is why they approached Yamaha to help them develop the AdLib Gold.
If they had some in-house engineers, they probably could have developed new cards much quicker, and might never even have needed Yamaha's help. They probably would also not have been as ambitious with the step from AdLib to AdLib Gold, but spread it out more, with yearly incremental product updates or such.

As for Gravis, I think what hurt them the most was the lack of proper OPL2/OPL3 compatibility. The GUS Extreme was a great card, but it was too late. If the original GUS was designed like that, they would have been far more competitive, and might have actually taken the market by storm (basically the same recipe as what made the Sound Blaster the standard: being 100% compatible with the current standard, and adding extra stuff).
Instead, they originally aimed at the GUS being an 'Audio Card Enhancer', assuming you'd already have an OPL2/OPL3 card. That made it not interesting enough for people who didn't have a sound card yet.

I think Creative was a simple Asian clone shop that accidentally struck gold with the Sound Blaster, and basically didn't really know how to step up from there. The original SB has a few obvious design flaws (such as the DSP not being able to play continuously), and design flaws plagued their products from there for many years (clicking and popping DSP on SB16, hanging note bugs etc).
They were a big name, and their products had big success in the market place, but the company behind it were still a bunch of amateurs.

http://scalibq.wordpress.com/just-keeping-it- … ro-programming/

Reply 122 of 129, by SirNickity

User metadata
Rank Oldbie
Rank
Oldbie
Scali wrote:

I think you are using the inverse meaning of 'prosumer' that I know.

This does actually explain a lot. Believe it or not, I've really tried to see this from your perspective (otherwise, what's the point of engaging in the argument at all?) I just haven't been able to square my experience with the SB Pro with how I perceived your expectations to be. This helps though. My definition of prosumer is more that I am not a professional, but I aspire to be. I would say I'm a prosumer musician. I have $1000s in rack synths, signal processing, instruments, software, etc. I don't make a dime on it. (Well, I've done paying gigs, but never the kind of thing where I contemplate leaving my day job.) Ergo, prosumer.

I would think of your definition as more "enthusiast". Willing to spend big(ger) bucks on high-end hardware, but not for anything but the enjoyment of it. When I really think about it, the difference between those two definitions is almost subtle enough to be meaningless. So, OK, I get it.

Scali wrote:

I mean, what's the point of all these people simply arguing: "I can't believe an SB Pro actually cost 600 guilders" or "Just because it says Pro and is twice as expensive as a high-end SVGA card doesn't mean you should expect it to be any good" etc.
... I thought it would change your perspective on what I said, and how reliable my account was, but no, the goalposts were just shifted).
... Don't argue about that stuff, it adds nothing but insult to me. But that is what you want, isn't it? Call me a liar? Paint me like some fool who thought he was getting professional quality equipment because it said 'Pro' on the box? That is what you're doing, and I'm fed up with it.

Brother... you've got to work on being able to argue a point without becoming emotionally attached to it. Frankly, it's a skill everyone needs to refine (including myself) and it's sorely lacking today, but I digress.... Anyway, I can't speak for others here, but I absolutely have no malice toward you. I definitely don't think you're a liar. I have no way, nor inclination, to verify what you saw half a lifetime ago. I thought I saw at some point somebody quoting a figure of $600 for a Pro. Dollars, not Guilders. That seemed ludicrous to me, and I figured it was a mistake, or a mis-remembering at least. When you and I arrived at a conversion rate to USD, placing the Pro at half that price, then ... alright, I can see that being the case somewhere. Certainly wasn't when I bought one, but I wasn't first in line either. At any rate, I don't see any goalposts. There's nothing to "prove" in this, it's just a discussion. We're not arguing facts, for the most part, just opinions.

Scali wrote:

If you ask me, AdLib failed because the AdLib Gold was too ambitious.

I agree, and the rest of your analysis as well. Sadly they were over their heads in the PC landscape as it became. Which is truly unfortunate, I would have liked to see them succeed. It was always going to be a short battle, though. After Win 95 came out, and most people had access to CD-quality DACs made by whomever, the differentiating factors would have become meaningless. At least after the short-lived wavetable craze, when games moved to pre-rendered audio and/or software mixing. I'm amazed and confounded that Creative is still in business today, when nobody needs a dedicated sound card anymore. (Except professionals of course.)

Gravis definitely had something special, but underestimated the importance of backward compatibility. It was painful enough to have a PAS16 without SB Pro compatibility. At least it had an SB 2.0 clone and OPL3 though. Without that, it's a one-trick pony. A cool trick, but only the one.

Scali wrote:

[ Creative ] were a big name, and their products had big success in the market place, but the company behind it were still a bunch of amateurs.

640K!enough wrote:

Creative was never a proactive company in the sound world; they were purely reactionary, preferring to sell the same over-priced products as long as they could, until it became clear that they had to do something if they didn't want to become irrelevant.

I disagree. Their pacing, I think, was on-point. Competition with, e.g., Media Vision certainly spurred them into action, but I think the fact that they always had an answer to that competition is exactly what made them successful. Of course you want to keep selling product X for as long as you can before X+1 is necessary. R&D is a fixed cost, subsidized by the number of units sold, right? Nonetheless, any time someone released something better, they had X+1 just around the corner. Never long enough to fall behind, which means they were anticipating it and preparing for it -- which, ultimately, is where AdLib failed.

640K!enough wrote:

For someone who claims not to be defending Creative, you're certainly acting like a devotee to the Church of Sim. Are you really trying to dismiss the historical record and pretend that Creative survived because they were the only company that understood how to price their products? If that's the case, I don't think there are any facts that anyone could present that would convince you otherwise.

Try me. I have my opinions, but I'm definitely willing to change them in light of new evidence. That's why I'm here in this thread. It's certainly not to change your mind, but if I poke holes in your argument, you will either fail to convince me, or you will succeed and I will learn something. Either way I'm good.

I'm just giving Creative credit where it's due. Yes, I have a special place in my heart for all things Sound Blaster. I have good memories, and now I have a humble collection of them. However, I was a big fan of Media Vision back in the day, I would love to own a GUS at some point, and I'm intrigued by the AdLib Gold. I totally agree that the SB line had a lot of flaws. A. LOT. Whew. I don't consider them amateurs, not even the engineers, I think they were just getting products out the door before they were totally and fully baked, and then it was.. eh.. good enough. Which it was. Not perfect, but good enough. Perfect is the enemy of good.

How has that changed with anyone else, though? Remember when BD vs. HD-DVD was a thing, and players started coming out with dual-format compatibility? Except it was largely broken, and they kept promising "it'll be fixed in a firmware update." Yeah right. How many of those saw even two updates past the factory version? That's just a product of capitalism.

Reply 123 of 129, by Scali

User metadata
Rank l33t
Rank
l33t
SirNickity wrote:

Brother... you've got to work on being able to argue a point without becoming emotionally attached to it.

There you go again, making this about me personally. I wanted to have a discussion about Creative's products, but everything I say turns out into people judging me, rather than discussing the actual subject.
I've given you many warnings already in this thread, but you keep doing it. What's worse, you don't seem to understand that you're doing it?
I mean, if you keep responding to people by personal insults, the obvious result is that people eventually get fed up with these personal attacks.
Trying to frame that as "becoming emotionally attached" just shows how little you understand.

Also you say you argue about opinions, not facts. But to me, the price of 600 guilders was a fact, yet there was a huge argument about that. Many other things I said were facts, not opinions. I simply wrote about my experiences with the SB Pro.

As for the term 'prosumer', you could have checked the interwebs to make sure you understood it properly, before attacking me. Wikipedia has a page on it:
https://en.wikipedia.org/wiki/Prosumer#Professional_consumer
And this definition is exactly the same as how I used the term. So basically I had to endure your shit because you have this weird idea of what a 'prosumer' is.

http://scalibq.wordpress.com/just-keeping-it- … ro-programming/

Reply 124 of 129, by Great Hierophant

User metadata
Rank l33t
Rank
l33t

The Sound Blaster was "good enough" for its time. Creative's sole moment of genius was to market a card that offered something to everyone at a reasonable price point. For many people, the Adlib compatibility and the Game Port were worth the price alone. Adding MIDI support, klunky though it was appealed to budding musicians. Digital sound capability had previously been the domain of high-priced DSPs, but Creative used a generic, not particularly complex microcontroller to let the card and the RAM communicate with each other. The SB's Digital Sound Processor's capabilities were not the killer feature at first because speech took up a lot of disk space, but the basic command structure was in place and by the time enhancements came, programmers would have been familiar already with Creative's DSP commands. I have not mentioned Game Blaster because no one really cared about it. After that they pretty much reacted to or bought out their competitors, Media Visison, Advanced Gravis, Ensoniq, Aureal.

http://nerdlypleasures.blogspot.com/ - Nerdly Pleasures - My Retro Gaming, Computing & Tech Blog

Reply 125 of 129, by SirNickity

User metadata
Rank Oldbie
Rank
Oldbie

Scali, dude... breathe, man! 😀 I'm not attacking you. I have never attacked you. I've never insulted you. I'm telling you, not for my defense, but for your own mental well-being, you're reading into this something that just isn't there. I have never demanded proof of your claims. I might've asked, rhetorically, in what universe an SB Pro cost so much -- but probably only when there was a misunderstanding about currency, and I would've asked the same question had I been staring at the price tag in the store. Incredulity, but not disbelief. I sincerely apologize if I came across as aggressive, but AFAICT, you're taking my words out of context of their intent, and that's no way to be looking for a fight where there is none.

The only bit about this any of this that has been personal is me trying to bring your head back down from the stratosphere. The rest -- is about a sound card. Why should that be stressful? You have no obligation to me to prove anything. I'm just talking with you. OK? Really.

That's all I can do. If that's not enough, I would invite you to feel free to endure no more of me by ignoring my posts. Last thing I want is to cause some bloke I don't even know to blow a gasket over trivial conversation. It's exhausting for everyone.

Reply 126 of 129, by Scali

User metadata
Rank l33t
Rank
l33t
Great Hierophant wrote:

I have not mentioned Game Blaster because no one really cared about it.

I think above all, you also need a bit of luck in this game.
If you take Tandy/PCjr sound, the chip is very simple. But since it's all the people had to work with, some composers really went out of their way to extract the maximum from the chip, getting some pretty nice results. My personal favourite is the Rob Hubbard-quartet of Kings of the Beach, 669 Attack Sub, One on One and Skate or Die.

I think if the Game Blaster had some 'killer app' titles like that, it could have been quite successful. The hardware is simple, but certainly more capable than Tandy/PCjr, and is also capable of sample playback.

http://scalibq.wordpress.com/just-keeping-it- … ro-programming/

Reply 127 of 129, by 640K!enough

User metadata
Rank Oldbie
Rank
Oldbie
Scali wrote:

I think if AdLib had a generic SB-clone in-between, a delay on the AdLib Gold would not have hurt them that much.

The biggest problem is that it was apparently a manufactured delay rooted in collusion. Creative desperately wanted to avoid having their "standard" supplanted by a more open, viable one, and repeatedly did whatever they could to undermine any such effort; see also VBE/AI.

Scali wrote:

But if I recall correctly, the real problem is that the AdLib company was not a company of technical people, but rather musicians/people with a vision. The AdLib card was designed for them under contract. So they could not design their own hardware in-house. Which is why they approached Yamaha to help them develop the AdLib Gold.
If they had some in-house engineers, they probably could have developed new cards much quicker, and might never even have needed Yamaha's help. They probably would also not have been as ambitious with the step from AdLib to AdLib Gold, but spread it out more, with yearly incremental product updates or such.

They do appear to have started that way. In fact, they weren't even really intending to start the PC sound revolution when they started; they wanted to produce software to teach music and give budding musicians some basic tools to get started. What became the MSC seems to have been a by-product of that effort. Realising what they had, they decided to re-brand themselves, modify the package and promote it as a product for general computer users, not just musicians. The brilliant move -- the one thing that allowed them to succeed where Creative had failed -- was hiring the QA/consultancy firm to help promote the card to developers. From there, it ended up at Sierra, support was added to King's Quest IV, and the PC audio revolution was born.

As time passed, they do seem to have added some of their own technical people. As mentioned, it wasn't an entirely in-house design, but as far as I'm aware, the card was not designed for them by Yamaha. They were collaborating on the Gold Sound Standard/Magic chipset, and that is likely where Creative found an opportunity to shove the wrench into the gears. Based on evidence, the control chip was produced for them by a separate ASIC sub-contractor. Someone correct me if I'm wrong, but I was under the impression that most of the hardware work was sub-contracted to another Quebec company: Lyrtech.

Scali wrote:

As for Gravis, I think what hurt them the most was the lack of proper OPL2/OPL3 compatibility. The GUS Extreme was a great card, but it was too late. If the original GUS was designed like that, they would have been far more competitive, and might have actually taken the market by storm (basically the same recipe as what made the Sound Blaster the standard: being 100% compatible with the current standard, and adding extra stuff).
Instead, they originally aimed at the GUS being an 'Audio Card Enhancer', assuming you'd already have an OPL2/OPL3 card. That made it not interesting enough for people who didn't have a sound card yet.

A big part of the problem for Gravis was the quality of software provided by their partner, Forte Technologies/e-Tek Labs. They often took many shortcuts, skipped any form of error-checking or reasonable sanity checks, etc. From the earliest UltraSound software to the very latest InterWave releases, it remains quite easy to find error conditions that simply aren't handled at all, and you're given no indication of what went wrong, or that an error even occurred. The design was also let down by the way the software handled low memory; the compression and instrument substitution really hurt its playback, compared to the competition.

Gravis really meant for the UltraSound to be "the future of sound and music on the PC"; it wasn't just an audio card enhancer, it was the new standard -- Creative Labs, Yamaha, eat their dust! Frankly, it showed definite promise, especially for 1992. Had they produced an OPL3 daughterboard and really refined SBOS (or stuck basic SB compatibility on an on-board MCU, as Ensoniq did, minus the wavetable-based FM emulation), they might have really had something. The ACE was a later, cost-reduced product, but was far from their flagship.

Their other fatal mistake was getting caught with their pants down when AMD bailed out of the audio IC market. That, I think, is what led to them reverting to the GF1 and producing/licensing the Extreme.

Great Hierophant wrote:

Creative's sole moment of genius was to market a card that offered something to everyone at a reasonable price point.

You're forgetting the essential ingredient: doing enough work to figure out that Ad Lib was using the YM3812 and also making it accessible at I/O address 388/9H. That compatibility is what allowed their card to get off the ground; it may otherwise have ended up as just another Game Blaster.

Reply 128 of 129, by rasz_pl

User metadata
Rank l33t
Rank
l33t
Great Hierophant wrote:

Digital sound capability had previously been the domain of high-priced DSPs, but Creative used a generic, not particularly complex microcontroller to let the card and the RAM communicate with each other. The SB's Digital Sound Processor's capabilities were not the killer feature at first because speech took up a lot of disk space, but the basic command structure was in place and by the time enhancements came, programmers would have been familiar already with Creative's DSP commands.

wait, wait, "The SB's Digital Sound Processor's capabilities" ?? were there any apart from trivial compression(basically a lookup table)?
Afaik Creative clever marketing called it DSP to scam people into thinking it was a real Digital Signal Processor (as in very fast, often vectorized, with hardware fixed point MAC opcodes, capable of actually useful things like FFTs)

Open Source AT&T Globalyst/NCR/FIC 486-GAC-2 proprietary Cache Module reproduction

Reply 129 of 129, by Scali

User metadata
Rank l33t
Rank
l33t
640K!enough wrote:

The biggest problem is that it was apparently a manufactured delay rooted in collusion. Creative desperately wanted to avoid having their "standard" supplanted by a more open, viable one, and repeatedly did whatever they could to undermine any such effort; see also VBE/AI.

What I tried to explain is that Creative could only do this because of the position AdLib were in.
Namely, the AdLib was introduced in 1987, and had not been changed/upgraded at all, until the AdLib Gold would relieve it in 1992.
That is 5 years without updates.
Had AdLib updated the card in-between, they could either have made their own Killer Card before Creative did... or they could have released an SB-compatible clone shortly after Creative did, and keep the advantage of their presence in the marketplace.
But now, they allowed Creative to take over the market with their SB, and the AdLib Gold was a do-or-die product: it had to be ready in time, and it had to be good. So Creative could just frustrate the release a bit, which was enough to kill the company, because the original AdLib was already wiped out by the SB in the marketplace.
That's my point: if AdLib had an updated product in-between, Creative could not have wiped them out that easily, because Creative simply could not have gained as much marketshare and built up their own brand as much.
5 years without product updates is where the real problem was. Not the AdLib Gold delay.

http://scalibq.wordpress.com/just-keeping-it- … ro-programming/