Kaide

The catch there is if you want to use the HDMI 3D spec, this is mostly true. If you just need to sync up the shutter glasses that came with the SMS, that’s another matter. It should work in the mode where the clock speed of the SG is adjusted without having to actually implement the 3D spec at all. Otherwise the screen tearing would not be nice. The lag the TV adds would probably desync the glasses though, but if you could add an adjustable delay to the signal for the glasses it could be overcome. This is probably the easiest answer, since it should be possible to create a delay circuit controlled by a potentiometer. You can also avoid needing a buffer while supporting the 3D HDMI spec by using the lower resolution side-by-side format versus the frame packed format BD uses. That would just need to be able to output “plain” 1080p30 to pull it off (PS3 does this). But you would need the HDMI scaler to support this form of frame packing directly, and that is something only Kevtris would be able to answer. And it would add a frame of lag as it needs to be fully buffered. This is the “cleanest”, but requires the TV to support 3D glasses, when newer ones don’t. Hah, you beat me.

I’m not really sure what point you are trying to get at exactly?

That is impressive. I think implementing it on the SNT depends on when the scaling happens. If it’s a post-step after the PPUs, then it gets a lot more complicated and expensive to implement. I suspect this is the case since the MSG has pretty much the same scaler with tweaks. It also saves some RAM if you only need to blit into a low res buffer and scale as part of outputting to HDMI. If I had to guess, this is done in BSNES by having the PPUs draw into the full-size buffer. Mode 7 wouldn’t need to scale, while sprites and other modes would do scaling as part of the blitting process into that buffer.

(Double Post)

What do you think is being “cached” beyond the SRAM? Settings have to be saved out manually to the built-in NAND that holds the firmware. It’s thrown away on power off. Same with the simulated SRAM. Everything else that would be loaded is R/O, or part of the system’s internal state that can be thrown away on power off. Honestly, power on/off isn’t instant because it is a momentary switch. It isn’t an on/off switch. Something has to poll/trigger off the switch being pressed to turn on or off. Latency isn’t exactly a high priority for such a circuit, and I’d imagine vampire draw is a bigger concern. The thing is, while your suggestion is prudent for many systems that run a full blown OS in them with HDDs/SSDs that have internal write caches, etc... here not so much. I think you’re reading more into this than there is.

With games that saved to SRAM (vs EEPROM or MRAM), losing power to the SRAM causes the save data to be lost. The battery is there to provide continuous power to the SRAM. Interrupting power to the console is not the problem, as the original consoles cut power to the PCB/cart when the power switch was opened. The issues you bring up happen in situations where there’s a cache, and the cache is then written out somewhere. Original carts don’t fall into this category. Instead, it is mostly more modern things like Everdrives, since the SRAM content has to be swapped out when switching ROMs. The JB firmware is probably similar, but I’m not sure when it moves the contents of SRAM to the SD Card. There’s no real reason the Mega SG or Super NT need to “cache” anything itself without the JB.

The idea is generally sound. I've yet to find a good implementation of it, but most of them are done in the analog domain. Doing it after the ADC might be more reliable, but in general I think the question is: how reliable is the noise floor, in practice? And is it possible to reliably detect the signal without introducing a delay before it "clicks" on? Generally the places I've seen it used fall down on the latter more than the former.

Yeah, that's the main draw for running the SNT/MSG in 720p for me on a 4K display. I don't care if the 4K display is doing NN or not, it's clear that doing it this way maintains the "correct" look of the scanlines better. Same with the SNT. And since I'm playing on an OLED, having a little letterboxing looks fine to me. ---- So a quick update, the replacement MSG is on its way back. And their logistics folks sent out the replacement via SmartPost instead of Next Day. So yeah, this thing has been shipped 3 times, via ground each time, each time taking about a week. Fun.

My guess is more that it’s because the Colecovision uses the Z80. Same CPU used by the Master System and included in the Genesis (which is how it gets backwards compatibility IIRC). It’s probably similar enough that it could use the existing Z80 implementation that was included for the Sega consoles. Not sure why other non-Nintendo Z80 consoles aren’t included though. As for a GBA FPGA, if Nintendo patented anything in the hardware you need to reimplement, then it gets murky. You can try to work around them, but you can also wait for them to fall into the public domain in a few years, probably (20 years on the patent IIRC).

This was the bit I was getting hung up on. I could swear on original hardware you didn’t need to do this part. Maybe my memory is bad. It’s been a while since the Sega CD was hooked up.

I might have to haul out my original again, but I’m sure at least the slowdowns happen on original hardware. Sonic 2 has some ugly slow down issues especially with rings.

I’ve gotten a return label now and shipping it back later today to get the version I ordered. No marks for not noticing until it was opened. And that is what I wanted to see, so it seems like things are getting through the pipe and they have a plan now rather than chaos they weren’t expecting to hit them. It’s being shipped Ground on their dime back to the logistics folks in Texas (mixed bag, unfortunately, since I’m not exactly close to Texas). Upside, return package should be overnight. So about a week turnaround for me. Here’s the thing, I can and will forgive mistakes, but it doesn’t mean I can’t facepalm at the decisions that lead to the mistakes, or compound them in the face of frustration. I also follow the “you get one” philosophy. So it’s not like I’m going to slag them in the media for this. I’d also like to point out to folks more generally there’s a difference between venting frustrations at those logistics folks, voicing a want to see certain improvements in their support story going forward (learn from this, Analogue), and conspiracies. It feels like in places folks piled on to people displaying all three of these, when the third one is the one that cannot produce useful feedback, while the first two are maybe uncomfortable, but damn useful when trying to run a business.

Yeah, the problem came up more when I responded to that asking what the exchange would entail, and getting a reply that didn’t answer the question, so I have to ask it again. That’s where the frustration begins to set in when it does take so long to per round trip.

It’d be nice if turnaround wasn’t 24 hours per email. They failed to answer my question about how the exchange would work from yesterday, so I’m waiting again to find out how the exchange will work. Or hell, just make it possible to have a real-time conversation so I can sort this out.

I’m still trying to figure out what an exchange with Analogue looks like, with their support. One person commented that if you opened it, no exchange for you. I’m trying to confirm that, because I didn’t even notice until the wrapping was off, because the packaging is so similar between the SKUs. So far, they don’t need help making themselves look bad, but this is more inexperience and incompetence in handling logistics/packaging than anything else, and then compounding slow/bizarre support on top.

I wonder how feasible it would be to do an FPGA + ODE where it exposes a USB port letting people use a CD/DVD drive or memory stick. That’s really what I’d be interested in.

Not at all. Other than a little leftover flux on two pins, they are all poking up, with no damaged pins like yours has.

I think you're right. Although in that case, I'd imagine that even if the tip of a pin breaks, it's still "good" as long as you get a good solder joint. Knowing that, Pin 7 looks pretty suspect now that I get a better look at it. It looks a bit like a bodge.

Yeah, those spikes are just solder. It happens when the solder "sticks" to the iron when it is pulled away from the joint. These are just pretty pronounced, and probably happened because it was soldering on an assembly line. The pins for the connector are surface mount, and don't go through the board. But that solder job is still noticeably worse than mine. So if you still have your iron, or know someone who does some soldering, it wouldn't hurt to reflow the pins.

I guess I’m not seeing the pin issue you are describing? I may be missing it. The solder job definitely looks like a somewhat wonky robotic job at temp settings that aren’t exact for the solder rather than a reflow or similar. The spikes from where the soldering iron lifted off the pad are on mine as well. If it is the solder job that is the problem, it could be fixable by pretty much anyone with an iron to reflow the pins.

The specs are likely leftover flux. Especially if they are using a lead free solder, going by what I see on my adapter. In the connector that you plug the MS cart into, all pins are present. The card edge connector for the Mega SG is clean and in good condition there too. If you have a missing pin, that’s likely your problem. I get what you are saying, but at the same time, I very much wanted the JPN buttons. I had a hard time picking what color I wanted for the Super NT, but not the SG. It’s not bad, but I just really like the red/blue mix. But I’m going to take the discount and make the EUR one a spare, or resell it for a small loss once I get the right one... Random question for you though, since you have the Everdrive X5 and the Sega CD. Do you have any problems using the Everdrive as a RAM cart on the SG? I can’t get it to work.

They could, but like zetastrike, I was offered a 2 week wait, or 25$ refund.

I’d be willing to compare photos of the problem areas with my adapter if it might help demonstrate your point to support.

Add me to the list of folks who ordered a JPN and got a EUR. Noticed it once the wrapping was off. Will see what Analogue Support says, but this isn't a great look if a bunch of folks are reporting it. All the shipping has been outsourced to a warehouse operator halfway across the country, and the boxes are almost identical except the bit of text on the side of the box. I'm willing to believe that it was a screw up, but it's not a great packaging job if your warehouse operator doesn't know how to tell your SKUs apart when shipping thousands of units.

Like with the Super NT not supporting external/expansion chips in the jailbreak, don’t expect it here. Both the CD and 32X rely on hardware outside the Genesis, and would require new cores added. In the case of 32X, I’m not even sure there’s room on the Cyclone V they use for the two CPUs in the 32X.