Jump to content


  • Content Count

  • Joined

  • Last visited

Everything posted by Nathan_A

  1. It sounds like the drivers posted earlier were a lost cause due to their resource forks being stripped?
  2. Other than the VRAM boards are their other physical hardware differences in the RAMDAC ("pixel") clocks to support those different resolutions, or is that something that's just controlled/configured by the firmware?
  3. Blargh! Womp, womp. Well, I'm glad it's going someplace that isn't the trash bin.
  4. Totally. It's even weirder to me because clearly there's some facility for a daughtercard on those things. I'm not sure what goes there. I think it's maybe for a SoundBlaster + MIDI/Gameport board kind of like some (all?) of the Houdini II cards have? However from looking at the Houdini II card I've got here, it's pretty clear that the little Vibra16 daughtercard isn't mechanically compatible with the sockets on that Reply board. I'm basically praying that the cabling I have for this Houdini II card will incidentally work on a Reply board like @new_user has, but I have no idea if it will actua
  5. Right? How nuts is that? edit: The 7100 version has it running along the bottom so that it has some clearance for the metal drive cage. The 8100 version has it running along the top so it can clear the plastic frame that stabilizes the HPV card.
  6. I would very much like to buy that card from you @new_user. I have a 7100/80 sitting right behind me literally waiting to have a card exactly like that one put in it. One of the crazy things about that card you've got is that it only physically fits in exactly two models of Mac, the 7100/66 and 7100/80. Electrically it should also work in an 8100, but the internals of the case design prevent it from going in because there's a flange in the case that holds the HPV graphics card in place which disrupts where that PCB protuberance on the card on the outside edge of the CPU socket woul
  7. Wow! Hello @t-shinba! It's really cool to see you here. Your product portfolio is incredibly impressive. Clearly a lot of care and attention to detail went into everything I could find online about your work. Hopefully we can revive some of it with your help. Cheers!
  8. That is an incredible bummer I was really looking forward to trying the card & drivers in my newly working 8100/100 this weekend before sending the Pickles down to @jeremywork.
  9. @pcamen sounds great. I'm sending my Pickles to @jeremywork this week.
  10. Put the machine all back together with the new PSU and she fired right up and booted into an install of OS 9 that the previous owner must have been running. It's basically a clean install except for having WeaveMaker Pro on it. Pretty cool vintage workstation.
  11. Does it change behavior if you choose other display resolutions and/or refresh rates?
  12. I'd certainly be curious to see what it looks like "in action" as it were.
  13. Well, good news. I acquired a spare power supply and when I plug it in and connect it to the mainboard the soft-power switch causes the PSU fan to start spinning. That's more than I could get to happen with the original PSU. I still haven't put the whole machine back together yet to see if everything functions or not because I have about two-dozen pieces of small plastic I need to figure out where they came from and whether or not I can super glue then back in place. So, that might be tonight's project. One thing that's weird with this machine is that the heatsink has a
  14. Is that garbled image still or do parts move, and if it's static are there clearly correlated parts if you drag something solid around? If the garbling moves around even while the onscreen image is just sitting there, then it seems like some kind of timing problem. If it's static and there's no correlated regions between groups of pixels when moving what should be logically correlated image bits around, then it could maybe be bad memory? If it's static, but clearly when you find a group of logically correlated pixels and you move some part of them around and their buddies move too
  15. @taddy interesting. I wonder what DIP switch settings you have to use in order to get the dynamic resolution switching to work? Maybe the DIP switch settings are just ignored on the XA and XA-Pro? @jeremywork in addition to comparing our respective ROMs we could probably also do a more fastidious physical examination of the cards for component placement and part numbers (though that would likely require prying off the heatsinks to be useful).
  16. I currently don't have any ROM dumping gear, but I'm more than happy to acquire some, since I'm sure I'm going to need it for the pile of things I have accumulated here. The best CRT I have handy to test with is a Dell P991. I don't think it'll go to the max resolution of the card, but it'll definitely exceed the limits of what the non-Pro version of the XA is supposed to be capable of, so we'd at least be able to verify that with extra RAM stacks the cards let you push into the list of resolutions and refresh rates that are only supposed to be available on the Pro-version. Whateve
  17. Well, @jeremywork what do you say? Shall we do a little bit of data collection & experimentation? I'm located in Portland, OR. Looks like you're down in SoCal someplace, so coordination and moving things back and forth should be too much of a hassle.
  18. That's what it looks like to me as well. It's possible they also have different pixel clocks for product differentiation, but I'd be really surprised to find out that they'd actually used physically different TI RAMDACs for that, but slightly less so to find out that such a thing might be controlled in ROM. The good news is that between a couple of us, we've got the gamut covered to find out, we just need a way to test and to get our RAM daughterboards to one or the other.
  19. Weirdly it seems like there are two different designs, though I don't know if they're any different fundamentally other than the connector types used. There's one set that are clearly stackable (e.g. like the ones on my card) and then another set that seems like they're supposed to be "caps" on the stack.
  20. Indeed. Though I expect real performance to lag behind physical refresh rate by a very significant margin. I'd like to get my hands on some additional pieces of gear from them (like the memory boards) as well.
  21. Whoa, that's incredible. Nice work! Though, it very much seems like there must be software to enable features beyond basic functionality. Because I'm definitely not getting QuickDraw acceleration. Or at least it doesn't feel like I am, and I'm only running it at 640x480 right now, which shouldn't be taxing the Nubus90 bus. I'm especially curious whether or not each card can do the same resolutions as the others given additional memory boards being stacked on. It seems plausible, but unlikely, that anyone would have bothered to make distinct versions of the card with di
  22. Picked up a bare ix3D on eBay for ~$30 after seeing your results here. Retail, not Apple, version. Looking forward to trying it out in-place of the Rage128 based card in my 400/1M G3 upgraded PowerCenter machine.
  23. Interesting. I wonder why the big disparity. It could also be differences in drivers. Both Number Nine and iXMicro had large variations in performance depending on driver versions. Curious to see what you find using MacBench 5. Thus far this is doing a pretty good job of motivating me to snag an ix3D or Revolution IV card and pair it with a 12MB Voodoo 2 card. As of last night, I've not had much luck getting the VGA-only PCI Voodoo 5 5500 working with a Mac ROM flashed to it. Part of me wonders if the Mac ROM defaults the primary output to be the DVI port that the Mac version of the card had,
  24. Geez, I didn't realize the first PowerTower's were Catalyst machines. PowerComputing sure got some mileage out of that architecture. I always preferred them to their Tsunami siblings for the decent on-board video and the faster bus speeds. They lacked memory interleaving, but that didn't seem to be much of a value add on machines of that era outside of niche circumstances. Otherwise, faster bus speeds tended to show obvious benefits. On my PowerCenter the VGA Enabler jumper is in the open position. Lastly, I don't think MacBench ever had 3D tests in it from what I can r
  25. The PowerTower machines are Tsunami architecture. The PowerCurve and PowerCenter machines are Catalyst. I don't know what their Pan & Zoom is for in the control panel. Kind of seems like an Accessibility thing. Sort of like Crtl-Scroll in OS X (I could be wrong about the key combo). I more meant that High-res desktop publishing in terms of synthetic benchmarks back in the day usually had to do with how much smarts there were in the ASIC to manipulate and shift around big buffers of data in and out of the actual framebuffer without the CPU needing to intervene to move memory aro
  • Create New...