Jump to content

DirtyHairy

Members
  • Content Count

    702
  • Joined

  • Last visited

Community Reputation

785 Excellent

3 Followers

About DirtyHairy

  • Rank
    Dragonstomper
  • Birthday 08/19/1980

Profile Information

  • Gender
    Male
  • Location
    Germany
  • Interests
    Music, Guitar, Coding, Retrocomputing & Emulation, Computers & Technology in general, Books, Food, Family

Recent Profile Visitors

8,991 profile views
  1. I wasn't thinking about a multimeter; there are meters that you plug in between USB port and power supply / hub and that measure current and power consumption over USB. As for the fan, I would do something like @SpiceWare suggested first, just to see whether this is really the issue. If it is, then it might be worthwhile to look for a more permanent solution. That's what's happening: threading is enabled, but the emulation itself isn't threaded, and that's what's loading the core. I am very sceptical about whether there is a way to thread ARM emulation --- each cycle of the CPU requires the previous cycle to complete, and I don't see a way to parallelise ARM and 6507 code either --- data for the ARM gets set up by the 6507, then the ARM runs while the 6507 is stopped on NOP, and then the 6507 runs again with the data the ARM calculated during its run.
  2. As the slowdown is intermittent, I have a vague idea that this may actually be thermal throttling. The NG firmware clocks the R77 at 1.2GHz, which is higher than the original clock speed of 1GHz, but still within the spec. However, even with the increased clock, CDF and DPC+ games drive one of the CPU cores at almost 100%. Maybe the activity of the USB bus increases power consumption of the SOC and increases the thermal load, causing the CPU to throttle every now and then in order to prevent thermal damage. Do you have equipment to measure the power consumption of the R77? It would be interesting to measure consumption with and without the daptors attached. If you feel adventurous, you could also open the casing and provide some additional cooling to the CPU (which only has a pretty small heatsink, see https://github.com/stella-emu/stella/wiki/Retron-77), like an external fan. Cheers -Christian
  3. I'm afraid we have found a few issues with Stella 6.3 that particularly affect the R77. Notably, savekey-enabled and CDF cartridges fail to work, and the dumper does not work correctly (aka 6.3 won't play real cartridges). In addition, the dumper may send Stella into a kill-restart loop. I've removed the build to keep everybody from finding out the hard way There'll be an updated R77 build soon.
  4. Yes, I think this is feasible, but you'll have to do some modifications. The R77 firmware is based on an old version of the Orange Pi uboot and kernel trees. The hardware is stripped down, though, and I am not sure whether there are any changes from hyperkin to the device tree. To my knowledge, the only things R77-specific are the dumper (which communcates with a MCU over tty) and a kernel driver that handles joysticks and hardware buttons (using GPIO and i2c). At a minimum, you'll have to remove the dumper from the init script and start Stella directly, and disabling the Hyperkin kernel driver is a good idea, too. Additional changes may be required in case there are more hardware differences, and you may have to reenable more OrangePI stuff in the kernel config. Provided some experience with (embedded) linux, I think it can be pulled off. I don't have time to do any actual works on this, but I'll be happy to give you advice
  5. The same here: do you still have plans for AtariVox support on the D9?
  6. I've uploaded a new build to github --- that one should be fine.
  7. Ups. That sounds like I built from an old tag. *slaps himself around a bit with a large trout* . I'll pull the binaries and do a new build.
  8. I have created the R77 build for Stella 6.3.0: https://github.com/DirtyHairy/r77-firmware-ng/releases/tag/stella-6.3.0 Have fun!
  9. You can definitely swap CPU and RIOT, and afaik you can swap the TIA too, you'll just get wrong colors.
  10. And another update: this fixes issues with some heavy sixers that would not previously run the Uno firmware.
  11. Now that's good news. Thanks alot for testing, and thanks alot to @batari for the hint about JAM --- that was the key. I'll release the fix as v17.
  12. You're right, that may really be what's happening. If the sixer comes out of reset while the Uno is still starting up and not emulating the firmware ROM (or running the dumper), then it could end up executing garbage from RIOT and TIA and hit a JAM.
×
×
  • Create New...