Following this article, Intel UK Press Relations summoned yours truly to a meeting so that he might help me to address 'inaccuracies' in that article.
Following this rushed spin meeting which Les Hewitt, a thermal engineer at Intel UK also attended, I remained generally unconvinced.
Intel seemed to become annoyed when the company line presented was being challenged, and seemed to refuse to accept the feedback we were presenting from some of Intel's customers.
Hewitt confirmed that BTX was in fact something that Intel had developed internally very soon after the introduction of its ATX standard - I pointed out that the www.formfactors.org website that Intel constantly referred to, was in fact an Intel owned website.
During the course of that meeting Intel categorically stated that the original article was wrong when we said: that "Taiwanese... ...manufacturers - and quite a lot of them it would seem - are being rather less than enthusiastic or co-operative, about the sweeping changes and support that Intel is asking, nay demanding, of them.
Our story also gave an insight when Intel was attempting to move the market to BTX.
The article said that: "Intel says that its customers should "commit" by May of this year, and secondly that it's Intel's desire to have 40% of all new systems Intel BTX ready by the end of the year". We questioned whether that was possible.
Fast forward to the halls at Computex Taipei today. It's June and we haven't yet found any BTX chassis that are in full-scale production with the major PC chassis manufacturers. What we did find was clear evidence of the very vocal opposition we reported earlier.
Chieftec Ind. Co. Ltd is one of the world's leading PC chassis manufacturers, and Jason Yang's business card says he's the CEO of Chieftec.
He referred to a morning meeting that had just been held with Intel, where it was confirmed that significant elements of the BTX form factor had still not been finalised.
One of the main things still not finalised is a metal plate which must sit below an Intel BTX form factor mainboard, and to which the massive processor heatsink for these BTX system bolts to. This, he pointed out, must be very strong indeed and would in itself cost $1US to supply let alone develop.
Metal plate which must sit below an Intel BTX form factor mainboard to which the massive BTX heatsink must be bolted.
Chieftec's CEO was clearly frustrated by Intel on a number of levels, and referred back to the mounting holes that had to be provided on an ATX chassis backplane to accept the 'through mainboard mounting scheme which Intel had insisted was part of its 'standard' at the time of introduction of Pentium 4. These, simple mounting holes, he claimed, cost time and money to implement yet Intel seemed oblivious to its partners' concerns.
I don't recall ever seeing this being adopted by any system builder, and all the production systems I have seen use retention mechanism and mountings that are supplied with retail boxed Intel Pentium 4 desktop processors.
Gigabyte were in fact a mainboard manufacturer that had also been caught out by believing that Intel would be sticking by what it had been saying about using the 'through mainboard' heatsink mounting scheme.
The general thrust of what Yang was saying was that this was another example of how Intel would basically make it up as it went along, and if it cost case manufacturers money, then it just didn't care.
The impression gained was that he simply wasn't happy with Intel and it's BTX plots.
He confirmed that he knew of no changes required by AMD to accommodate any thermal or acoustic concerns with its processors.
He said that it cost around $250,000 US to produce the tooling for an Intel BTX chassis and it even cost circa $10,000 US to produce a single, laser cut prototype. Each time Intel came forward with a new proposal this was the cost implication to each chassis manufacturer. This was confirmed by Enshen Huang, the General Manager of IN WIN Development in the Netherlands, another long established major PC chassis manufacturer.
At the meeting with Intel earlier in the year it maintained that BTX would be cheaper to produce than ATX although, it conceded, not at first. Intel customers I have spoken to sure didn't seem to see things quite the same way as Intel, and you can see why.
Across their entire ranges of chassis, both Chieftec and IN WIN were only displaying one prototype each of a BTX chassis and Enshen Huang said that if BTX was to come, it would be very late into Q4 and more than likely next year. Chieftec's CEO seemed more assertive, saying that if it was to happen then 2005 was the most likely time. I haven't even been around to see all the heatsink, PSU and mainboard manufacturers yet. µ
small>Chietec's BTX prototype - please note we're told that Intels' BTX design is not yet finalised
It's time for our regular two-step through the Google news
Bug bounty offer: accepted