There was a G5-compatible PCIe X1900 GT, and its ROM is on The Mac Elite. As far as I know, nobody has gotten the AGP version working. The X1300 is, in theory, 1/4 of an X1900. So as a prerequisite, if the AGP X1900 could be made to work, then maybe the X1300 PCI could be made to work. Each is a...
Thanks guys, yeah, having some fairly low-stakes fun with these old PCI cards.
I finished up jt's VisionTek card today. It got a recap, and I replaced the diodes with STTH310s, which are 1.7V forward voltage. I measured 843mV across them while the card was running, so they're definitely doing...
There are no failure-prone RIFA safety caps in the SE power supply or on its analog board. You were having some display issues before this happened, right? I suspect the flyback transformer is what failed.
The SE and SE/30 power supplies are compatible, but you may need an analog board repair...
You can get a SWIM that will work in your SE from Kero's Mac Mods in Japan for $48: https://en.infinityproducts.co.jp/product-page/super-wh-iwm-swapper-plcc44-to-dip28-swim-adapter
For an extra $16, you can also get flashed ROMs, but they're not pin-compatible, so there's a little extra work...
I had some NTD40N03s on hand, so I tried swapping them in for the NTD14N03s. As expected, no improvement. Still burning hot. Here's the reference schematic for this regulator:
The VisionTek card doesn't implement it exactly as shown above, but it shares one thing in common that I hadn't really...
The limited experimentation I've done with bus overclocking suggests that EDO modules can work at faster bus speeds (up to 60MHz) where my FPM modules do not.
I have been misunderstanding all of the TLV431-based regulators on these cards. They are not switchers at all. While it is possible to build a switching regulator with a shunt reference like the TLV431, the configuration here is a linear regulator. They find a gate voltage where the FET's RDSon...
Somewhat unexpectedly, this saga continues. @Trash80toHP_Mini asked me to convert a PCI 9250 for him, and I agreed to take a shot at it. Specifically one of these, a VisionTek with a DMS-59 connector:
The DMS-59 output is odd and was clearly chosen for its ability to fit into a low-profile...
Funny that this thread is still getting posts. The OP has not been here in over a month. In perspective, I would say that his software was a great success.
Glider up to at least version 4 is playable with music on a 68000 in b&w. It will bog down a bit when there are a multiple sprites on the screen, but it's playable.
PCI Vendor 1668 is Actiontec, whose bridge chip comes up in reference to the Sonnet Tempo Trio card. If that's what you have, very nice score.
You could open the case and look.
Edit: I think the reference to the Sonnet Tempo Trio I found in this thread may be a red herring. Actiontec makes...
I played around with Rs and Cs in LTSpice because it's a lot faster than trial and error with a soldering iron, and settled on 10 ohms and 4.7uF. That gives me the attenuation I want (down to <20mV) without dropping too much voltage:
I had a 4.7uF part on hand; granted, in an 0805 package...
Funny, when researching the LTC6905 and jitter, I came across a few references to the Digilent BASYS2, which is a Xilinx Spartan 3 FPGA dev board: https://tilde.ini.uzh.ch/~tobi/wiki/doku.php?id=dig:fpga2#faq
Apparently they had the same problem! Nice to know I'm not the only one.
It finally occurred to me to try measuring the peak-to-peak noise on the 5V input to the oscillator, and look at the FFT:
I can't go >50MHz with the FFT on this scope so I might be missing some interesting things going on at higher frequencies, but as far as I can tell, I'm dealing with about...
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.