Jump to content

kevtris

+AtariAge Subscriber
  • Content Count

    859
  • Joined

  • Last visited

  • Days Won

    20

Everything posted by kevtris

  1. I heard about it, but I would have to get ahold of the repro to test it out.
  2. This is correct. The look is softer but it's very very slight and you might not even notice it except by toggling it on/off. I personally leave horizontal interpolation on.
  3. If you use your own usb cable, make sure it is a good one. Cheaper cables won't have big enough wire on the power, and it will cause excessive voltage drop. One of the crappy cables I used originally during testing had over 1V of drop at 500ma.
  4. I highly suggest powering it down between game changes, just like a regular system. Your save games might not live through the experience. The giant bombcast guys lost their metroid save hot-swapping the cartridge on that stream. (and I made a comment to this effect in the chat on said stream)
  5. I think it's mainly to reduce confusion and sources of issues.
  6. This isn't generic advice. I wrote it, and this is how it works. You cannot flash a "bad" firmware, there are multiple checks for proper firmware, and even if somehow you managed it, it will gracefully recover when you try again. If you turn it off mid-update it will detect this on the next power up and flash the LED waiting for you to put an SD card with firmware on it in the slot. It's "unbrickable" via the update mechanism.
  7. This is correct- it looks for the firmware file only. It must be in the root directory, though. Note that it will look for the *first* firmware it sees- If you have 2 or more files on there, it probably won't load the one you are expecting. (technically it searches through and will load the first one in the directory list, so most likely it will load the first one you copied over.) So just delete all firmwares except the one you wish to load. There's no other restrictions. In fact, I used my sd2snes SD card to do updates too. (just to be clear: obviously you must put the SD card into the system for the update to work, it won't find it on the sd2snes!).
  8. You can, but it will slightly raise the noise floor on the audio. I doubt most people will even notice so you can leave it enabled if you like. The passthrough audio would work, but I can't support pass through video- that'd need a lot of stuff- a video ADC and then composite video decoding, etc. The one that just has a video output will work though (on a cable or whatever, I am not sure how that works). It'd just be using the snt as a power supply and controllers. Sort of like that "super 8" thing did with NES/famicom games. You'd plug your AV cable into it, and then there was a little cable that came off of it and plugged into the NES.
  9. The power is basically directly connected to the port at all times that it is plugged in, so there's no way to disable it in the menu. Put a very small square of tape over the LED to cover it up if it bothers you. Power draw is very low and around 10-20ma so it is basically enough power to light an LED.
  10. The PAL modes assume the same resolution as NTSC to start, 256*240 pixels or so. The black bars / vblank is not sampled or upscaled, only the actual active display area. Yes, PAL games will work in 5x on NTSC mode, but they will run 20% faster (and might crash / look funny if they assume they have 70 lines of vblank to work with, but are only given 21). If you want to confuse your drunk friends, slip them PAL super mario bros running on NTSC. It's hard to tell at first what's wrong. Mario feels like he's got ballbearings in his feet. It makes completing jumps and stuff a bit trickier.
  11. This isn't true. the PAL and NTSC NES has the exact same output resolution. The reason is what GH said above though- I don't have enough blockrams on the FPGA to buffer enough scanlines to do 5x for PAL. The scanlines come in much faster than they can be displayed, because vblank on PAL is a lot longer (70 scanlines or so) vs. NTSC (20ish). The HDMI adjusts for 50Hz by making hblank super duper long, so vblank is about the same length on 50fps vs. 60fps. PAL/NTSC scanline rate is very similar, as in total scanlines per second. There's just more of them per frame in PAL (and on NES, the "extras" are just blank lines, so this means PAL NES's should letterbox somewhat I believe on a CRT). HDMI compensated by slowing down the scanline rate per second for 50fps vs. 60fps, hence the rub. If they had simply increased vblank time it would have been a non-issue. but nooOOOooo they had to lengthen each scanline, instead :-)
  12. Correct me if I'm wrong, but you have yet to post pictures/video of what it is doing, and pictures of the cable (showing soldering work). It's impossible to diagnose the problem. Also, the AVS doesn't even output analog signals and is HDMI only. Still not sure why you are trying to run composite into a scaler to get HDMI, vs. using the HDMI the nt mini outputs directly. I mean, you're free to do so but it doesn't make much sense to me vs. using native output directly. Be that as it may, it should work OK no matter what. There's plenty of video/images of the analog output quality from the nt mini on the internet so I find it hard to believe it's an nt mini issue.
  13. I highly recommend against doing this. This will backfeed the power supply (i.e. 7805) and can cause chips to self-destruct.
  14. re: LED brightness. I have added it as an option now, so you can set it from off to full on in 32 steps. This allows one to select any LED brightness they like.
  15. This is correct. Anyone selling them for $250 was scalping. I don't doubt that happened on the first batch or two since they sold out so fast, and I do not condone it. The first bunch were preinstalled so we could work out all the kinks and make sure that when others started doing installs that it'd go smoothly.
  16. yep, that's the one. it is fairly thorough. I like how it will keep looping until it fails or you turn it off, so you can keep it running for a few hours.
  17. I am not bragging; I am stating the realities of how much money it cost to get them done. I wished I could've made the boards for $5 a pop, that'd been awesome. People don't realize how much investment you need to get stuff made in any great quantity. If your doodad costs you $30 in parts (note: the hi def is significantly more than this) and you wish to make 500 of them, that's a $15000 up-front cost that you need to spend before you see anything back. There is plenty of risk involved- will the boards work when you get them back? will they use legit parts or pirate parts? will they just take the cash and run? will there be major hardware issues that didn't show up on 5 units but do on 500, causing expensive rework or outright rejection of the entire batch? You are paying for the investment in both time and money that it took to get it produced and available for sale. If you can do it for less, by all means. The margin I make on the boards is getting close to the point where I might not bother to make any more, since my time is worth more being spent doing dev work for a paying customer for the time invested in the project.
  18. It's the one with the most extensive tests. there's several of them floating around; the one I like has the big laundry list of things it tests, then does the ppu graphics demo. I don't know why the adapter wouldn't work, and haven't tried the test suite.
  19. the bike uses the expansion port so nope. I have a sufami turbo; works fine. the rifle is light gun based probably so nope, and the burn in cart works fine (I used that a lot for testing)
  20. It can't... directly. however you can make/load a custom LED pattern and set it to anything you like from off to darker or whatever. For solid black just make a file that's 768 bytes long, full of 0's. The file format is just RGBRGBRGB byte triples and you can force it to display one of the RGB entries specifically, or cycle through them. This is true for the nt mini, as well. The patterns on that work on this.
  21. You're comparing something made in batches of 500 by a private individual vs. something made in 1 million or more quantities by a multinational organization which huge buying power. Unfortunately I am paying 5-10x more for parts and assembly than a sony or nintendo would and that is reflected in the price. I didn't think people would be so interested in the hi def, but I am glad they are and thank them for it. You could say the same about certain NES carts that sell for $1000-$10K or more. They can be played on a flash cart, but people like collecting and will pay top dollar for the uber rare.
  22. It can't, but I have a special PPU development board that plugs into the cartridge port and I can run real PPUs + the FPGA PPU in parallel while games and test programs run. This lets me do direct A:B comparisons in real time. At the very start, I did use an nt mini with a PPU board though until the super nt hardware was ready to go. I am not sure what the problem is, but it sounds like cable issues. If you made your own cable, it's possible the ground isn't connected to ground or is open circuit or similar. An open ground will result in a very poor picture since the video is being grounded through other means (or not at all). The composite output in NES mode is exactly the same, voltage wise, as a stock system. I tested it here during development on two PVMs, and on several other TVs here and at a friend's house. The sillyscope, earm, oscilloscope shows no noise or other anomalies on the DAC output lines either. If you made your own cable, make sure you have the proper output mode pins grounded or else it might be outputting RGB or component instead.
  23. Seeing how the FPGA and transmitter chip alone total around $20, not including the rest of the parts, soldering, testing, programming, shipping, etc. your estimate is off by over 10x. If you can produce a hi def NES mod for $5, by all means go for it. Not sure why you have such a vendetta against me, but I will play along in case this isn't a trollolololol. You have absolutely no idea how much money and time I have invested into this project. Countless hours and around $10K of tools, test equipment, and funds went into the first 2 rounds of prototypes. After that, I sent a new car's worth of money to China to get the first run of production boards made. The amount of money I tied up in the boards is substantial. A lesser person might have resorted to crowdfunding, but I put most of my savings into getting the first run made. Your $100 water bottle argument is pretty specious; no one needs high resolution graphics to live. This is what we call a "discretionary expense". If you think you can design, program, and produce a comparable product for less, by all means do so. I sure wished I was making a $100 profit on each unit, that'd be awesome. Unfortunately for me it's less than half this. On the upshot, I did pay off my initial investment and threw some of it into my mortgage principle so there is that. There is a "long tail" of supports and fixes for products like this, even mature ones. This means that I am still putting in unpaid hours on improvements and fixes to the project as time goes on. This is fine, and is just part of the deal and I am not complaining, just laying out some of the finer details people might miss. edit: forgot to mention. The hi def is going to be back in stock in a week or so. I ran out last year and couldn't get more made because I was deep in the super nt project, and couldn't spare the time to program/test them and get them produced. I received more boards a few weeks ago and programmed/tested a bunch so it's ready to go again. Hopefully I won't run out again next time and can order before that happens depending on what paid work/projects I have going.
×
×
  • Create New...