VOGONS

Common searches


Reply 60 of 139, by Tetrium

User metadata
Rank l33t++
Rank
l33t++

One year later 🤣! 😢
Actually 2 years later 😲

The suspense will make me age faster, please hurry up 🤣

Whats missing in your collections?
My retro rigs (old topic)
Interesting Vogons threads (links to Vogonswiki)
Report spammers here!

Reply 61 of 139, by Beegle

User metadata
Rank Member
Rank
Member
Great Hierophant wrote:

Nice card, and just to think that I used to work right next door to such an historic place in Boston.

What's funny, and they detail this in the interview, the address in Boston was merely a PO Box - intended for the American market to feel "safer" when dealing with AdLib.

hard1k wrote:

Any news so far?

Project is going slower than expected and took a long time to get over the first major hurdle. I had severe problems with the audio data which took me 6 months to resolve. Basically since I recorded with 4 different pieces of equipment, all their internal clocks weren't going at the exact same speed. When I tried to overlay the different sources, and aligning the starts, at the end of the interview there was a +- 2 second offset in the different tracks because of a problem known as drifting. Drifting of this importance isn't acceptable and creates a weird robotic effect on voices when present - I definitely had to fix this issue before moving forward.

640K!enough wrote:

Very interesting! That's guaranteed to have some great value for the nostalgic AdLib fans. If you haven't already completed too much of it, I hope you are not planning to overlay a translation over the original voices. Was there any discussion of successor MediaTrix Peripherals? I know that Mr. Chalifour did some work for them, too.

Thank you. Not planning for an audio dub, but subtitles instead.
And yes, Mediatrix was discussed, however in less detail.

Tetrium wrote:

One year later 🤣! 😢 Actually 2 years later 😲 The suspense will make me age faster, please hurry up 🤣

Calm down, we're only 7 months later at the moment!
My first contact with Henri Chalifour was indeed in 2015, but the actual interview was only done last December 😉

So, some news.
I managed to fix the audio problems which prevented me from moving forward with the editing process. This took me 6 months to find, and then apply the tedious fix.
Now the actual editing has begun! Out of the 4.5 hours of interview, I made cuts every time the discussion changed to a new subject. In essence, that means every clip of video now discusses a single topic. Clips range from 10 seconds to a few minutes.

I have then rearranged the "useful" clips together, and left the "useless" clips on the side.
Useless clips could be someone thinking for too long, someone saying a joke, me rephrasing a question, or people searching through documentation, etc.
The useful clips have been placed in a more or less "logical and semi-chronological" order to have a clearer progression in the story of AdLib. That took a while (1 month) and now this part is complete. All together, the pertinent interview bits cover 2.5 hours of dialogue, which I'll combine, cut, rearrange to mimic a steady discussion as much as possible.

Here's a pic of all the clips together :

montage screenshot.JPG
Filename
montage screenshot.JPG
File size
251.57 KiB
Views
2724 views
File license
Fair use/fair dealing exception

Editors may notice that I only have one video track, even though I had 3 cameras on set.
I'm doing the edit in a manner that will be coherent on the audio side first (so the discussion flows correctly), then I'll re-separate the video tracks in three in order to make the visual edit based on the timing set by the audio track. Some time in the future I will have to shoot "B-Roll" videos to spice up the visuals.

So the project is still moving forward rest assured, but it's taking a long time. It's the biggest film project I've edited so far, and it makes me feel small. I like a challenge though, and I hope to deliver a good quality regardless of the time it will take 😀

The more sound cards, the better.
AdLib documentary : Official Thread
Youtube Channel : The Sound Card Database

Reply 63 of 139, by hard1k

User metadata
Rank Oldbie
Rank
Oldbie

Absolute coolness, Sir 😀

Fortex, the A3D & XG/OPL3 accelerator (Vortex 2 + YMF744 combo sound card)
AWE64 Legacy
Please have a look at my wishlist (hosted on Amibay)

Reply 64 of 139, by Beegle

User metadata
Rank Member
Rank
Member
firage wrote:

I hope you're only cutting out bad jokes. 😀

Yes, and keeping some good ones 😉

hard1k wrote:

Absolute coolness, Sir 😀

Thanks!

The more sound cards, the better.
AdLib documentary : Official Thread
Youtube Channel : The Sound Card Database

Reply 65 of 139, by 640K!enough

User metadata
Rank Oldbie
Rank
Oldbie

There is one thing I'm hoping you touched on in the interview: I keep reading that AdLib had no engineering talent and sub-contracted all design. Allegedly, the MSC came from now-defunct Quebec company Lyrtech, and the Gold was all Yamaha's doing. This doesn't sound right at all, but can you put that to rest once and for all?

Reply 66 of 139, by Beegle

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

There is one thing I'm hoping you touched on in the interview: I keep reading that AdLib had no engineering talent and sub-contracted all design. Allegedly, the MSC came from now-defunct Quebec company Lyrtech, and the Gold was all Yamaha's doing. This doesn't sound right at all, but can you put that to rest once and for all?

It's mentioned in the interview, regarding the Gold, that the development of the card was done by Lyrtech, because AdLib didn't develop in-house.
However, the point that remains unclear is whether Lyrtech developed the card alone, or cooperated with / relied on Yamaha for some aspects.

I'm pondering doing an interview with more Adlib/Lyrtech employees and merge those bits into the interview.

The more sound cards, the better.
AdLib documentary : Official Thread
Youtube Channel : The Sound Card Database

Reply 67 of 139, by 640K!enough

User metadata
Rank Oldbie
Rank
Oldbie
Beegle wrote:

It's mentioned in the interview, regarding the Gold, that the development of the card was done by Lyrtech, because AdLib didn't develop in-house.
However, the point that remains unclear is whether Lyrtech developed the card alone, or cooperated with / relied on Yamaha for some aspects.

So there is some truth to that, after all, though we still don't know if Yamaha had a hand in it. That leads me to wonder who developed their software. It seems fairly obvious to me that it was done in Quebec, since some of the sample code I managed to find (with AdLib copyright notices, variously written as "AdLib" and "Ad Lib") contains comments in that distinctive Quebec style of French, mixed in with largely English documentation.

Beegle wrote:

I'm pondering doing an interview with more Adlib/Lyrtech employees and merge those bits into the interview.

This is getting to be quite the production, isn't it? 😁 This has always been one of the areas of technology that I thought left many questions, and any effort that tries to shed light on it is a worthwhile one, in my mind.

Reply 68 of 139, by Mike_Rosoft

User metadata
Rank Newbie
Rank
Newbie
640K!enough wrote:
So there is some truth to that, after all, though we still don't know if Yamaha had a hand in it. That leads me to wonder who d […]
Show full quote
Beegle wrote:

It's mentioned in the interview, regarding the Gold, that the development of the card was done by Lyrtech, because AdLib didn't develop in-house.
However, the point that remains unclear is whether Lyrtech developed the card alone, or cooperated with / relied on Yamaha for some aspects.

So there is some truth to that, after all, though we still don't know if Yamaha had a hand in it. That leads me to wonder who developed their software. It seems fairly obvious to me that it was done in Quebec, since some of the sample code I managed to find (with AdLib copyright notices, variously written as "AdLib" and "Ad Lib") contains comments in that distinctive Quebec style of French, mixed in with largely English documentation.

Beegle wrote:

I'm pondering doing an interview with more Adlib/Lyrtech employees and merge those bits into the interview.

This is getting to be quite the production, isn't it? 😁 This has always been one of the areas of technology that I thought left many questions, and any effort that tries to shed light on it is a worthwhile one, in my mind.

I may be able to help with the information to the Gold 1000 and 2000 both products I was familiar with as I was bought into Ad Lib Multimedia to see what was purchased by the European buyers... and then design the ASB and MSC product line.. so I lived in it for a few years ...

Reply 69 of 139, by Beegle

User metadata
Rank Member
Rank
Member
Mike_Rosoft wrote:
640K!enough wrote:
So there is some truth to that, after all, though we still don't know if Yamaha had a hand in it. That leads me to wonder who d […]
Show full quote
Beegle wrote:

It's mentioned in the interview, regarding the Gold, that the development of the card was done by Lyrtech, because AdLib didn't develop in-house.
However, the point that remains unclear is whether Lyrtech developed the card alone, or cooperated with / relied on Yamaha for some aspects.

So there is some truth to that, after all, though we still don't know if Yamaha had a hand in it. That leads me to wonder who developed their software. It seems fairly obvious to me that it was done in Quebec, since some of the sample code I managed to find (with AdLib copyright notices, variously written as "AdLib" and "Ad Lib") contains comments in that distinctive Quebec style of French, mixed in with largely English documentation.

Beegle wrote:

I'm pondering doing an interview with more Adlib/Lyrtech employees and merge those bits into the interview.

This is getting to be quite the production, isn't it? 😁 This has always been one of the areas of technology that I thought left many questions, and any effort that tries to shed light on it is a worthwhile one, in my mind.

I may be able to help with the information to the Gold 1000 and 2000 both products I was familiar with as I was bought into Ad Lib Multimedia to see what was purchased by the European buyers... and then design the ASB and MSC product line.. so I lived in it for a few years ...

I've sent you an email.

The more sound cards, the better.
AdLib documentary : Official Thread
Youtube Channel : The Sound Card Database

Reply 71 of 139, by Beegle

User metadata
Rank Member
Rank
Member

Jade Falcon, there is a huge amount of progress, but it's still low-level, so hard to really show off any visible advancement from a viewer's perspective.
Here's the timeline at the moment - I'm rearranging the clips into "clusters" that talk about a same subject, and making the clips flow into one another like there was no cut.

molasses.JPG
Filename
molasses.JPG
File size
102.63 KiB
Views
2397 views
File license
Fair use/fair dealing exception

Because here's an example :
The interview is 4 hours long.
We spoke about subject X at minutes 21, 1h15, 1h30, 3h05 and 3h44.
I cut all these moments, and stuck them side by side.
And now I rearrange them so they make sense.
I remove all the hesitations, stuttering, searching, duplicate explanations.
And I stitch the clips together so it seems that the person spoke about the subject in one go without interruption.
The end result of this little group of clips is maybe 2 minutes long.
I have done ~20 of them, still have about ~60 to do.
Doing one of these clusters of 1 subject, takes about 2 to 3 hours of editing to make them presentable. So still ~150 hrs to go for this step.
This is taking forever. Kill me please.
The real editing of the interview will be so much easier than this step.

In the meantime here's something to keep you busy.
It's a late development Adlib Gold prototype. Note the different circuitry top-left, at some point there was apparently a SCSI connector on there.
And surprisingly, the prototype card was... 16 bit.

GBP_2271_smol.jpg
Filename
GBP_2271_smol.jpg
File size
1.18 MiB
Views
2397 views
File license
Fair use/fair dealing exception

@edit : the card is not mine, although I'd love to have it 😉

Last edited by Beegle on 2017-10-09, 14:07. Edited 1 time in total.

The more sound cards, the better.
AdLib documentary : Official Thread
Youtube Channel : The Sound Card Database

Reply 72 of 139, by hard1k

User metadata
Rank Oldbie
Rank
Oldbie

Oh my. If you ever ever ever consider parting with it, please let me know, we'll work it out.

Fortex, the A3D & XG/OPL3 accelerator (Vortex 2 + YMF744 combo sound card)
AWE64 Legacy
Please have a look at my wishlist (hosted on Amibay)

Reply 73 of 139, by Jade Falcon

User metadata
Rank BANNED
Rank
BANNED

Nah, I understand. Just hopes its good 😀 better yet perfect 😎

Edit. Nice card. Like tge presudo 16bit slot. Whats the back of tge card look like?

Reply 74 of 139, by Great Hierophant

User metadata
Rank l33t
Rank
l33t
Beegle wrote:
In the meantime here's something to keep you busy. It's a late development Adlib Gold prototype. Note the different circuitry t […]
Show full quote

In the meantime here's something to keep you busy.
It's a late development Adlib Gold prototype. Note the different circuitry top-left, at some point there was apparently a SCSI connector on there.
And surprisingly, the prototype card was... 16 bit.

GBP_2271_smol.jpg

Interesting, it looks like they tried to copy the Sound Blaster Pro's 16-bit-ness. In other words, the 16-bit features are limited to extra IRQ and DMA selections and entirely optional.

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

Reply 75 of 139, by Beegle

User metadata
Rank Member
Rank
Member
Great Hierophant wrote:

Interesting, it looks like they tried to copy the Sound Blaster Pro's 16-bit-ness.

However, the Gold was in developed in response to the original SB from 1989; and by the Pro's announcement in Summer 1991 the Gold was probably already shown with prototypes like these in trade shows.
So my guess is that they used a 16-bit controller for SCSI? And when the option was removed, the +8 bits were left unused.
I don't have a picture of the back, but it lost 6 connectors.

I want to buy the card.

The card isn't mine, but Raymond's! And I doubt he wants to sell it so don't go and try to find his phone and address now.
This is now clarified in the post.

Jade Falcon wrote:

Nah, I understand. Just hopes its good 😀 better yet perfect 😎
Edit. Nice card. Like tge presudo 16bit slot. Whats the back of tge card look like?

I sadly didn't take a pic of the back. But this prototype revision lost 6 connectors when going from 16bit to 8 bit.
And the interview certainly won't be perfect. But I'm trying to make it as polished as possible and it takes a while.

The more sound cards, the better.
AdLib documentary : Official Thread
Youtube Channel : The Sound Card Database

Reply 76 of 139, by 640K!enough

User metadata
Rank Oldbie
Rank
Oldbie
Great Hierophant wrote:

Interesting, it looks like they tried to copy the Sound Blaster Pro's 16-bit-ness. In other words, the 16-bit features are limited to extra IRQ and DMA selections and entirely optional.

This is a frequent misconception that I have seen. The Ad Lib Gold was well on its way to being on store shelves around the time of the Sound Blaster Pro; it seems unlikely that Ad Lib would have had time to "copy" anything. There was even an official (pre-bankruptcy) brochure that I had from Ad Lib that compared the Ad Lib Gold 2000 to the Sound Blaster Pro. The only item where the Pro had an advantage was in the number of FM channels it offered; 22 due to the dual YM3812 design, versus 20 for the Gold's YMF262. The shipping Gold 1000 seems to have made the SCSI feature optional. It doesn't look like the 2000 ever got into the retail channel, even though the software appears to have been ready for it.

If you search, you can probably find old reviews where the only Creative offering was the Sound Blaster Pro, and the reviewers criticised Creative for taking so long to get an OPL3 card out the door. If you actually look at the specifications, the Gold was actually more than just competitive with the as-yet-unannounced Sound Blaster 16, and, though this is quite subjective, offered generally superior sound quality. If allegations are true, it's no surprise that Creative was worried enough to engage in anti-competitive behaviour with Yamaha to ensure that the MMA chip "wasn't passing testing" until it was too late.

Reply 77 of 139, by 640K!enough

User metadata
Rank Oldbie
Rank
Oldbie
Beegle wrote:

In the meantime here's something to keep you busy.
It's a late development Adlib Gold prototype. Note the different circuitry top-left, at some point there was apparently a SCSI connector on there.
And surprisingly, the prototype card was... 16 bit.

Nice card indeed. It seems to have a pre-production control chip and YMZ263 (MMA) on it, too. That looks more like a prototype of what would become known as the Gold 2000. I have never read anything about it actually using the AT portion of the slot for data transfer, just availability of additional IRQ and DMA lines.

Interestingly, the CD audio module has "Ad Lib Multimedia" etched into it.

I also recently read an interview with Mr. Deschênes, formerly of Ad Lib Inc., in which he claims that Ad Lib approached Yamaha to specifically request the YMF262 design and feature set, as a compromise between cost and sound quality. That leads me to wonder if they had to pay for that as a custom development project.

I also notice that there are different markings on the chip in the same position as the YAC512 on the shipping Gold 1000; more pre-production ICs?

Mike_Rosoft wrote:

I may be able to help with the information to the Gold 1000 and 2000 both products I was familiar with as I was bought into Ad Lib Multimedia to see what was purchased by the European buyers... and then design the ASB and MSC product line.. so I lived in it for a few years ...

Can you give a little more background on how you fit into the story? Were you sent to Quebec, by whom?

Reply 78 of 139, by Jepael

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

I also notice that there are different markings on the chip in the same position as the YAC512 on the shipping Gold 1000; more pre-production ICs?

No, that's basically (older?) compatible YM3015 (at least I think it reads Y3015 on the photo) which is also a DAC but it needs higher supply voltage. YAC512 is a (newer?) 5V device recommended for OPL3.

640K!enough wrote:

The only item where the Pro had an advantage was in the number of FM channels it offered; 22 due to the dual YM3812 design, versus 20 for the Gold's YMF262.

That would be correct from a certain viewpoint (marketing?) when number of channels are calculated in a way that tries to maximize it even if it was a corner case nobody really uses.

A single YM3812 has 18 operators, that are used as nine two-operator channels, or if rhythm mode is enabled, then you only get six two-operator channels and five channels for rhythm, totaling 11 channels.
So fair enough, with two chips, you do get up to 22 channels in total, but they are divided between the left chip and right chip, so you have 11 channels on left and 11 channels on right.
To produce a tone from both speakers, you need one channel from both chips. This is also how it makes the backward compatibility with single chip mono designs, write to standard FM addresses actually writes to both chips. Granted, when controlling both chips separately, you can decide if one, both or neither are in rhythm mode. I am not 100% sure but don't think it was possible to use the SB Pro mixer to combine these left+right channels into mono, they were one chip for left and one chip for right only.

The YMF262 on the other hand, has 36 operators in total. In OPL2 compatibility mode it boots into, it is practically 100% software compatible with single OPL2 chip.
But in OPL3 mode, it acts almost like dual OPL2 would. You get eighteen two-operator melodic channels, but, you can select if you send the channel output to left, right or both channels simultaneously.
So for playing mono music, you really have double the channels available when compared to SB Pro.
In rhythm mode, you get fifteen two-operator channels and five rhythm channels, which totals to the 20 advertised channels. So only one "half" of the chip can be set into rhythm mode. Because of this, it can't emulate being two OPL2 chips. But if rhythm mode is not used, under software control, it can act just like two OPL2 chips, one panned hard left and another panned hard right. It just requires that music drivers exist for both OPL3 and dual OPL2, or single driver can handle both cases.

That still leaves out other OPL3 features, like 4 new waveforms for more complex sounds, and combining up two-operator channels to get up to six four-operator channels for even more complex sounds.
Plus OPL3 has a faster internal processing of IO writes so music driver using OPL3 can waste less time waiting between OPL3 chip writes.

Reply 79 of 139, by Beegle

User metadata
Rank Member
Rank
Member
640K!enough wrote:
Mike_Rosoft wrote:

I may be able to help with the information to the Gold 1000 and 2000 both products I was familiar with as I was bought into Ad Lib Multimedia to see what was purchased by the European buyers... and then design the ASB and MSC product line.. so I lived in it for a few years ...

Can you give a little more background on how you fit into the story? Were you sent to Quebec, by whom?

I've sent this gentlemen an email 2 weeks ago. Got a short answer.
then I've sent a more detailed email with many questions. No answer...

The important thing is that Mike_Rosoft raises a few good questions that I have to investigate. And it seems I won't have a choice to interview people from pre-AdLib after all.
More work 🤣 yay! And more info about the Mélocapteur and Exercette products.

The more sound cards, the better.
AdLib documentary : Official Thread
Youtube Channel : The Sound Card Database