Found DirectX Time Capsule
I’ve been doing a little spring cleaning in my garage and naturally ran across a huge collection of DirectX memorabilia that I haven’t seen in many years. One particular gem was a silver case marked with a holographic DirectX logo that I don’t recall ever having opened previously. I cracked it open on my kitchen counter and realized that I had found a complete DirectX time capsule from March 1997. It was a press kit that Microsoft’s PR team had put together for a press event on Microsoft’s Media strategy. It sparked many memories of events that I had long forgotten. Recall that this case was prepared just three months before I was terminated and two months before the media strategy day I organized. This was a time capsule straight from the Talisman era. A brief examination of old emails suggests that this was a PR box intended for the launch of DirectX 5.0 at CGDC in April, 1997.
Inside the case I found a really neat collection of items, including an early drawing pad and several aluminum sheets of metal separating thin sheets of tracing paper. There was also a DirectX branded mouse pad. Great collectables, but most “interesting” was what I found in the PR folder.
The scanned contents of the PR folder on the far right are linked at the bottom of this blog. Ignore the folder marked “top secret” in the upper left corner… that’s for a future blog…
If you have been following my DirectX chronology, this was the era in which, after years of media strategy failures, Microsoft suddenly had a hit in DirectX. Following the amazing success of DirectX, a battle began for control over it resulting in the ownership of DirectX being handed over to a group from Microsoft Research whose own efforts in related areas had been less productive. Many groups with random multimedia initiatives of their own strove to attach the popular “Direct” brand name to their API’s. The result was an explosion of “Direct” API’s from random Microsoft groups wanting to piggyback their “less useful” media API efforts on the popularity of the DirectX name… few of which, beyond the original DirectX API’s, survived into the modern era. Microsoft’s answer to Adobe Flash, SilverLight also dates from this era by a different name… see if you can spot it in the sea of PR documents included here.
The time capsule also includes a reference to the mythical DirectX 4.0 release. *Note the true story of the missing DirectX version is that DirectX was one of the first Microsoft technology products to be built using a primitive version of Agile Project management. Now ubiquitous in the game industry and at Microsoft, Agile was little known in that era. DirectX was built in many parallel sprints with the final features of each version getting locked down as the release date approached. If a DirectX feature wasn’t going to be done in time, it was simply rolled into the next DirectX release scheduled a few months later. DirectX 4.0 was coming up short on features in time for its release date. We were faced with a choice between shipping a DirectX version that was light on new features or slipping our release date for the API which would have ended up close to Christmas 1996. Generally nobody was starting new game projects during that time of year so Eric, Craig and I made the call to just cancel the DirectX 4.0 release and bump its feature set into the DirectX 5.0 release scheduled to go beta at GDC 1997. It was that simple.
The time capsule also contains a lot of references to Microsoft’s “Active” platform in that era. The name ActiveX came from a desire of other groups to call their non-media related API’s “Direct” something. In an effort to try to preserve the “DirectX” brand as a gaming API, I was chartered with stamping out non-gaming uses of the DirectX name by other groups at Microsoft. In the case of the Internet Explorer there was a desire to call the browser plugin architecture “DirectInternet” or “DirectPlugin”, etc. I “influenced” the marketing team to think of a cool “X” brand name for internet related API’s that wasn’t “Direct”. They chose “ActiveX” from a long list of suggested names I submitted.
Although it has been many years I recognize the handiwork of Jason Robar on this case and suspect that he had something to do with its design and the many supporting developer quotes. Microsoft PR would not have been able to assemble such a kit without his assistance. I also recognize my own writing style in some of the white papers but I suspect that these documents are derivatives of white papers I had written at some earlier point.
This case is a monument to Microsoft’s desperate attempt to rationalize many disparate technology initiatives as some kind of unified whole comprising a grand vision and strategy that they wanted the press to accept and promote to developers. As I’ve mentioned before, I was under pressure from many directions to sell all this stuff to the game developers who were eagerly adopting DirectX and I had responded by pushing back HARD on these groups to worry less about justifying their waste of time projects and more about devoting themselves to building products somebody actually wanted. It was not a popular position but I didn’t care.
I would like to claim some moral superiority for the vision behind DirectX but I’m embarrassed to confess that even Eric, Craig and I felt compelled to bow to some of Microsoft’s crazy platform initiatives from that era. Direct3D would never have been necessary to create if it weren’t for the inter-group rivalry between Windows95 and WindowsNT that caused the NT group to refuse to share their OpenGL license with the Windows95 team. Making DirectX a COM (Common Object Model) API was another architectural abomination that we swallowed with some bitterness. COM was a strategy cooked up by the Office Products group to “unify” all of the Windows Office products in a way that made them highly interdependent without it being easy for them to be accused of engaging in “monopolistic” practices. The COM model had been foisted off on both OS groups as a grand strategy that everybody had to support. Every Windows API had to conform to the slow, confusing and cumbersome COM model. Half of the Developer Relations Group was devoted to promoting it to developers. We got a lot of grief from game developers for it over the years which I believe was rightly deserved. Here the time capsule shows Microsoft proudly touting the brilliance of COM integration across the DirectX family of API’s.
You can also see Microsoft’s schizophrenic positioning of OpenGL versus Direct3D support, trying to rationalize to the press that support for the two competing API’s was an intentional “strategy” rather than the internal political brawl that had spilled loudly into the public forum that it really was.
I think one of the things that is exciting about this box is that it also marks an exciting boundary line from the years that Eric Engstrom, Craig Eisler and I spent creating DirectX, to our efforts to lead Microsoft’s multimedia strategy onto the Internet. By this time senior executives within what had been the Windows 95 team who had once counted us as genuine nuisances had come to the conclusion that we were, in fact, powerful assets. For all of our crazy shenanigans we had ended up making them look brilliant and their opponents look ponderous and out of touch. Brad Silverberg who had formerly been responsible for leading the development of Windows 95 had been given the charter to dominate the internet with Internet Explorer and Craig, Eric and I had all been given front row seats in the initiative to do that. I have avoided chronicling a lot of the Internet era of media at Microsoft out of a desire to fully chronicle the creation of DirectX to the best of my recollection first but those stories are approaching a culmination soon with the collapse of the Judgment Day II event.
____________________________________________
From: Philip Taylor
Sent: Thursday, April 24, 1997 10:08 PM
To: Morris Beton MM – Games
Subject: FW: OpenGL vs Direct3D at CGDC
Importance: High
fyi, this is what amounts to the dev team rude q & a re OGL and D3D.
it actually tracks quite closely to our rude q & a, ie there is a place for both.
the additional comment re screensavers actually points to the fact that quake does not run on OGL, it runs on a very small subset hacked by id specifically for quake. On the net you can see similar comments, ie dont install the opengl32.dll that comes with quakegl for 3dfx or else your true OGL apps and screensavers will cease to function. repeat – quake does not run on OGL, it runs on a very small, special purpose subset
—–Original Message—–
From: Nicholas Wilt
Sent: Thursday, April 24, 1997 9:43 PM
To: Manhattan Project Development Team
Subject: OpenGL vs Direct3D at CGDC
Importance: High
This is bound to be a hot topic at CGDC. If someone sees you are from Microsoft and collars you about OpenGL vs. Direct3D, I recommend the following:
– Emphasize the positioning that Microsoft has always hewn to: OpenGL is for technical applications, Direct3D is for interactive ones
– If they persist, respectfully point out that OpenGL has no apps. glquake doesn’t count. If they bring it up, ask if they’ve seen it running on anything other than a 3Dfx? And have they noticed that the OpenGL screen savers crash on glquake systems?
–Nick
Leave a Reply