The real unicorn: a working 40 MHz VLB bus

[GiGaBiTe]

I never said 66 MHz host capable controllers were not common in later model machines. All cards had to indicate they could run at 66 MHz since the PCI clock was singular and provided by the host. As soon as you insert a card that doesn't advertise 66 MHz operation either via the OD mode pin or the config bit, the host controller keeps the bus clock at 33 MHz.

While this may be the case, server boards usually had multiple PCI bus controllers to avoid this. I used to have a Super Micro X5DPL-iGM which had three PCI buses, one for the 133 MHz slot, one for the two 100 MHz slots and one for the slower 33 MHz slots.

I had one quad Itanium server years ago which every PCI-X slot had its own bus controller so each card could run at whatever speed it desired without interfering with neighboring cards.

You are very much mistaken. PCI/PCI-X only operated at 33 MHz and 66 MHz. The 533 you are referencing is the MB/s (64 bits * 66.66 MHz / 8 bits per byte). I've designed a *LOT* of PCI cards in my day and even a few host controllers in Verilog and can practically recite the damned spec.

You are very wrong here, PCI-X went up to 533 MHz, for a maximum throughput of 4,266 MB/s.

https://en.wikipedia.org/wiki/PCI-X#Versions
https://web.archive.org/web/20100622025532/http://www.pcisig.com/news_room/faqs/faq_20/

 

[eeguru]

I stand corrected.

 

[dieymir]

40Mhz VLB was anything but rare. This works fine as long as you use late chipsets and VLB cards. I have an UMC U5SX40 running with a Trident TGUI9400Cxi and a Promise DC4030 VLB cards and the system runs perfectly without any WS.

50Mhz VLB is the real unicorn ...

 

[Stinkpot]

50Mhz VLB is the real unicorn ...

Cool, I have a unicorn. Not sure if it's running at zero WS or not...