I got an opportunity to talk with MIke Fulton, an ex employee of Atari Corp in the Jaguars mainstream lifecycle. He was the Developer Support Engineer Pre-Jaguar and then he was the Jaguar Tools Manager.
I shared with him different stories we had heard through the years about Atari's lack of communication with its developers. Here I am posting a brief summary of some of the anecdotes and his replies.
Mike Fulton wrote:
Chris, you've got a lot of information but you're missing perspective on a lot of these things. You're looking back on them years and years later and of course certain things seem more obvious in hindsight.
I wondered to him why it seemed Atari did not communicate with its developers better and work with developers to get the bug workarounds from those who had discoverd them for the Jaguars hardware, such as the UART bug. I relayed one story where Scatologic/4play had said when they reported to Atari they had found a workaround for the UART bug they claimed Atari treated them like they were incompetent and did not know what they were talking about. They related there was no real effort to negotiate to use their UART bug workaround.*
I don't recall who those developers were, but one thing you should keep in mind is that it was very possible for developers to present a list of symptoms as part of a bug report and for that list to be incomplete or contain a few extraneous items that would confuse the issue. It's not at all unlikely that whatever they told Atari simply didn't line up with what Atari knew, or thought they knew, about the problem.
Also, it was not uncommon for developers to occasionally get in contact with some Atari engineer who played no official role in providing developer support. Those engineers would have no idea of the process we would normally follow, and in many cases the results of those conversations were never passed on to the support people, even in summary. And once having made such contacts, those developers would forever expect those people to act in a support role, completely bypassing the regular support channels.
I can understand why developers would have wanted to talk directly to those people, but that probably caused more communications problems than anything else.
>>they claim Atari wanted them to send it over so they could evaluate it or some such and took it to mean Atari may have wanted to copy the code and use it without due credit or compensation.*
I don't recall this specifically so perhaps it happened after I left Atari[Mike left Atari around sept. 95], but here's what it sounds like to me: Some developer figured out a workaround, and Atari asked them to share it, and they didn't. You say they claim Atari didn't offer them anything for it, and that's probably true, but did they ask?
Robinson Requiem guy: My only contact with Atari was a producer who had no idea how the machine worked...
The contact information for Atari's developer support people was provided in the first few pages of the main documentation, and it included our phone numbers and email addresses. If this guy thought his only contact was the producer, he either didn't look at the documentation very hard or he didn't ask the producer.
Producers would occasionally come to developer support and pass along technical questions from their developers. Unless it was something pretty simple like "tell them to download the update from the BBS" we would get in touch with the developer directly when that occurred, rather than expecting the producer to pass something back.
I talked to him about the errors in the technical docs.
I would agree that the documentation could have been better in many ways. More inline sample code, for examples, would have been great.
We tried to incorporate corrections from developers into the documentation where we could, but those reports had to be independently verified and sometimes that took time. Someone who reported a typo and expected to get back a fixed version of the docs in a few weeks was not being realistic. We released new versions of the documentation periodically, but undoubtedly not as often as some would have liked (including myself).
Also, we discovered there were many cases where the developer who ended up working with the Jag dev system and documentation wasn't the same guy who we sent everything to in the first place. When we sent out updates, they'd go to the original recipient and didn't always end up making it to the new guy.
When we found out about these situations, we'd try to add the new people to our database, but some undoubtedly fell through the cracks.
I proposed what he appropriately termed crowd-sourcing for the problems. More open communication more sharing of problems with such things as the risc compiler.
I disagree. The RISC compiler was not ready for distribution. It was not even minimally functional by any reasonable standard. If we had sent it out, today you'd be asking us why Atari was sending out broken development tools.
Seriously, can you imagine: "Hey developers, here's a new compiler for the RISC processors, but you can only use it for code that compiles to less than 2K and doesn't compare values against each other."
Please... developers would stormed Atari HQ and made the crowd outside Frankenstein's castle look like a garden party.
It's romantic in hindsight to think that the problems could have been crowd-sourced away, but the reality is that if we had sent it out as-is, it would have been nothing more than a time-wasting distraction for almost every single developer who looked at it.
Let me anticipate a question: "So why did you give HVS the compiler, then?"
We didn't. They got it directly from Brainstorm. BS occasionally did stuff like that, despite our asking them not to any number of times. In this particular case, it turned out well, but more often than not it simply caused confusion because developers would end up reporting bugs related to versions of the tools that Atari hadn't even seen yet.
I asked him why Ken Rose did not know about the compiler and he was an in-house developer who now builds and targets compilers for a living. He also had lots to say about having to work completely in asm and the lost money and man hours that cost Atari. I reasoned that Ken Rose would of went through the problem like a hot knife through butter.
As to Ken's not knowing about the RISC compiler at the time, that's entirely likely. I don't know how many more times I need to repeat the idea that the RISC compiler was an uncompleted project in the very early stages and not ready for distribution. Ken was working on a specific game project. There was no reason he would have been made aware of new tools before they were ready for distribution.
Eric Smith knew about the issues with the RISC compiler at some point. I saw Eric on a daily basis and we talked all the time. Furthermore, those issues also affected assembly programming, just in a different way.
You keep talking about the situation like we had the "broken RISC compiler" hidden in a drawer for a year or two, but from the start of the project to the point where figuring out a workaround for the hardware problems became any sort of bottleneck was really no more than a couple of months at most. We didn't pull people like Eric Smith off their other projects because we were paying Brainstorm, who were very competent, to work on this stuff.
In *HINDSIGHT*, I would say that it's possible if Eric been pulled off his other projects (multiple) and been tasked to work on the compiler stuff, he might have been able to come up with something. The same is probably true of Ken Rose and a number of other engineers. But the conventional wisdom within Atari at the time was that there wasn't any way to create a reliable workaround for some of these issues. The fact that we had Brainstorm working on the project in the first place was something of a "Hail Mary" effort, hoping they'd be able to come up with something nobody else had thought of.
High Voltage Software being able to come up with what they did was totally unexpected. Whatever hopes we'd earlier had for a miracle solution had gone away by then.
Thank you for your time Mike. I just wanted to share with you some of the stories we have heard and take this opportunity to communicate what we the public have been hearing about for 20 years.
We're every bit as much as part of the "public" as you are, and we've heard all this stuff many times before. Most of the time we (or I) simply don't bother responding because it really, really, really doesn't matter any more.
I also asked him about MK3 and what he knew about it. He says he knew nothing about it.
*Going by memory here and summarizing. If my anecdotes are incorrect please get hold of me and I can retract/correct the statements.