NESRGB board available now

[ms06fz]

I don't see anything... Where's the problem?

[RGB32E]

I don't see anything... Where's the problem?

darcagn's pictures don't make it very "plain" You have to zoom into the picture to see the diagonal lines - faint as they may be in that example.

[ms06fz]

I don't see anything... Where's the problem?

darcagn's pictures don't make it very "plain" You have to zoom into the picture to see the diagonal lines - faint as they may be in that example.

Ah, I see. I think they didn't show up on my phone screen.

[Skips]

I found a possible issue with the new NES RGB revision and was wondering if anyone else that has received a new kit could try it as well. This problem is kinda like the problem we saw in the second run of the PCB's.

In Mario 2 the background changes on the character select when you select toad or Luigi. It seems to change to either white, brown, purple, blue, or white and is completely random what color it changes to. It does this on both Mario 2 and Mario USA. I have tried both the real cartridges and the Everdrive n8. I have not found this issue in any other game yet although I have not sat down and played through a ton of games yet. It is not the mod work, everything has a solid connection and everything works so far except this one part of Mario 2. I played through the first level and the problem still only showed up at the character select. It is not the PPU, it worked perfectly fine before sticking it in the kit.

Here is a super shitty and shaky cell phone camera video of the issue: https://www.dropbox.com/s/07al3fc60sygq ... 182820.mp4

**edit** I forgot to mention it does this in any palette.

And here are some SUPER shitty pictures!

[Voultar]

I'm betting it's an issue with v1.5 firmware, rather than the hardware itself.

If nobody else can chime in, and you have the resources, why not flash 1.4 to the board and see if the results are consistent?

[Pasky]

Looks like the exact same issue I had with FF1 when I entered a battle that was supposedly an 'installation error'

It was a firmware issue.

[Skips]

I'm betting it's an issue with v1.5 firmware, rather than the hardware itself.

If nobody else can chime in, and you have the resources, why not flash 1.4 to the board and see if the results are consistent?

Unfortunately no programmer. I ordered one today so I wont know for a bit. If anyone that does get their kit soon could try this to confirm its not just this one kit doing it that would be great.

[Voultar]

I'll bump one of my boards up to 1.5, to see if I can yield these same results.

Be back in a flash. (pun)

[eightbitminiboss]

Definitely not reproducible in 1.3.

[Voultar]

Yeah, it's in v1.5

Evidently ORB3D compatibility is breaking support elsewhere. lol

[TheRetromancer]

Yeah, it's in v1.5

Evidently ORB3D compatibility is breaking support elsewhere. lol

Oh, irony.

[Voultar]

I don't know if any of you other peeps are doing this. But tacking in headers to the JTAG interface on the underside of the NESRGB work great for fast programming.

[ms06fz]

Hm, so should I just order myself a USB blaster (clone) now, then? Is the clone hardware decent? I sure hope so 'cause the genuine article is crazy-expensive...

I don't think there are any more outstanding problems. This is most likely the final update.

OK, I'm a programmer myself so I know how it goes with bugs. I mean, it happens to all of us... But I just wanted to take a moment to appreciate this line.

[darcagn]

I don't know if any of you other peeps are doing this. But tacking in headers to the JTAG interface on the underside of the NESRGB work great for fast programming.

At this point, I just went ahead and did this, and I'll just keep it permanent, haha (and I still have the plastic cover, so it still looks clean normally):



And I did just upgrade to v1.5, and the issue has appeared in Super Mario Bros. 2 (USA). It only happens during the character select screen, however, so it's not a huge bug. But of course there is always the concern that other ROMs are affected in other areas. Doki Doki Panic uses a different character select screen and is unaffected, but Super Mario Bros. USA (the Famicom port of the US's Super Mario Bros. 2) uses the same character select screen and is affected.

Hm, so should I just order myself a USB blaster (clone) now, then? Is the clone hardware decent? I sure hope so 'cause the genuine article is crazy-expensive...

Yes, the clone device does exactly what you need, and that's what Tim bought and shipped out to customers when the first firmware problem happened.

[kamiboy]

Could someone let Tim know before he ships out more of the new boards?

[lettuce]

Ah crap, i ordered some more nesrgb's on Wednesday i bet they've been shipped out now, i was going to send the USB blaster off to another UK'er to use but i guess im going to need to hold on to if for a while longer

On another note, what is regarded the best sync pad to use, PPUV, CS, or V??

[viletim]

I found a possible issue with the new NES RGB revision and was wondering if anyone else that has received a new kit could try it as well. This problem is kinda like the problem we saw in the second run of the PCB's.

Yes, it's another timing related bug. Sorry, it's embarrassing there bugs like that still in there. This one has been there from the start. It's always been marginal but it seems that when I made changes elsewhere it changed the timing for one signal just enough. I don't think it will be a show stopper like last time.

It can be fixed by upgrading to version 1.6
http://etim.net.au/nesrgb/background_fault/

I found that when I placed a cro probe on /RAM_WE line the probem went away. A capacitor of 22pF does the job as well. The capacitor is soldered between pin 48 (third from the end) and ground.

The board pictured is one of my dev boards, that's why it looks so messy.

Or just leave it as it is unless some significant, game breaking problem is found.

Most orders from the US distributor have been shipped already so they will have version 1.5 software. No orders from me have been sent yet (I'll do it Tues/Wed) so they will all go out with version 1.6.

[kamiboy]

Great to hear. Keep up the good work. You are doing gods work.

[lettuce]

Most orders from the US distributor have been shipped already so they will have version 1.5 software. No orders from me have been sent yet (I'll do it Tues/Wed) so they will all go out with version 1.6.

What about UK orders Tim?

[darcagn]

Most orders from the US distributor have been shipped already so they will have version 1.5 software. No orders from me have been sent yet (I'll do it Tues/Wed) so they will all go out with version 1.6.

What about UK orders Tim?

Tim only has a special distributor for the US, he ships out the rest.

[The_Atomik_Punk!]

I found a possible issue with the new NES RGB revision and was wondering if anyone else that has received a new kit could try it as well. This problem is kinda like the problem we saw in the second run of the PCB's.

Yes, it's another timing related bug. Sorry, it's embarrassing there bugs like that still in there.

With all due respect viletim, this is exactly why I previously delayed ordering a NES RGB kit. I missed out on the first batch, and had the opportunity to buy a second batch kit, but trusted my gut and waited it out in case there might of been lingering issues (was on the Gametech_US group buy list). For me, that was certainly the right choice, as the whole debacle with the second batch would have been more of a hassle then it's worth, as I need to hire a modder to do the installation for me. I once again delayed ordering on this third batch, just in case something was overlooked again, and sure enough it was. My order would have been placed via the US distributor, and I would have had to send it back/ contract someone with a JTAG reprogrammer to fix it for me. I'm once again glad I didn't buy your product.

I understand this may not be a big issue to those with the requisite technical experience to fix the problem on their end, but for a regular consumer who's interested in purchasing your amazing NES RGB kit, this is not acceptable to have a timing issue that once again requires the customer to fix the problem. I realize that this product exists solely in the realm of the modding enthusiasts and not "laymen" (and that this only affects the US distributor's units), but I would rather wait and have a product that functions as it was intended, rather than rush to purchase one with a few kinks still to iron out. I was going to contact skips to see if he could mod my NES toaster with a RGB kits from this new batch, but looks like I'll be waiting for the next batch again, as he's the very person that brought this issue to our attention.

Please don't take my point harshly, I'm merely saying from a business standpoint (as well as customer relations angle) that it is preferable to take more time to ensure an optimal product is released, than to ship one with lingering issues expediently. Please keep up the great work Tim, I eagerly await the progress of your 4th batch!

[ms06fz]

You should bear in mind the kind of effort it would realistically take to catch a defect like this pre-release.
First off, there's no way to fully automate a test. Even if you automated the controller inputs you would still have to have a set of human eyes watching the game to see if it performs correctly.
Second, a failure could crop up in any game, and there are hundreds of NES and Famicom games out there. In this case it was an easy-to-reach part of a very common game, but it could as easily be buried deep in an obscure corner case of an uncommon game. To have a realistic expectation of catching a bug like this, you'd have to go through a lot of games, probably over 20 hours of testing for a decent regression suite. 20 hours of sitting around, watching two game screens to see if they still match, every time there's a new revision. And that still might not be enough to find a problem before the board ships.
So for a small operation like this, comprehensive regression testing may simply not be a viable option.

[bobrocks95]

With all due respect viletim, this is exactly why I previously delayed ordering a NES RGB kit. I missed out on the first batch, and had the opportunity to buy a second batch kit, but trusted my gut and waited it out in case there might of been lingering issues (was on the Gametech_US group buy list). For me, that was certainly the right choice, as the whole debacle with the second batch would have been more of a hassle then it's worth, as I need to hire a modder to do the installation for me. I once again delayed ordering on this third batch, just in case something was overlooked again, and sure enough it was. My order would have been placed via the US distributor, and I would have had to send it back/ contract someone with a JTAG reprogrammer to fix it for me. I'm once again glad I didn't buy your product.

I understand this may not be a big issue to those with the requisite technical experience to fix the problem on their end, but for a regular consumer who's interested in purchasing your amazing NES RGB kit, this is not acceptable to have a timing issue that once again requires the customer to fix the problem. I realize that this product exists solely in the realm of the modding enthusiasts and not "laymen" (and that this only affects the US distributor's units), but I would rather wait and have a product that functions as it was intended, rather than rush to purchase one with a few kinks still to iron out. I was going to contact skips to see if he could mod my NES toaster with a RGB kits from this new batch, but looks like I'll be waiting for the next batch again, as he's the very person that brought this issue to our attention.

Please don't take my point harshly, I'm merely saying from a business standpoint (as well as customer relations angle) that it is preferable to take more time to ensure an optimal product is released, than to ship one with lingering issues expediently. Please keep up the great work Tim, I eagerly await the progress of your 4th batch!

As a programmer myself, I can tell you that pretty much nothing will ever be completely bug-free. There will always be some obscure Famicom pirate or fan game/translation, or even something official that goes untested for a long time, that will reveal some incredibly minute bug in the software. If you're sitting around waiting for the one, final, ABSOLUTELY PERFECT batch, it's not happening. You'll have to decide at some point that you can live with very small bugs that don't affect you very much, or just not get an NESRGB, that's all there is to it.

[Skips]

The addition to the through hole where the Famicom audio channels get mixed in is super nice. I was able to use a connector for that as well because of it. I can now remove kits from the AV Famicom without having to desolder anything. Works super nice if you have to test stuff a lot since it allows you to quickly disconnect it.

So far the bug with the 1.5 firmware is not a huge issue, Mario 2 and Doki Doki Panic seems to be the only games I can find an issue with.

**Edit** I am closed to commissioned mod work until after July. Sorry guys but I am out (what little I do have left has been spoken for by others already) of wire, cable, connectors, and all the other miscellaneous parts needed to do this stuff. I wont be ordering more parts/components for modding until August. I do this stuff as a hobby not a business so I don't keep a large stock like some modders do. What little I do have gets used on personal projects and the occasional commission.

**DOUBLE EDIT** I found the issue in 1.5 in Metroid as well when going from the blue tunnel to the gold hallway (the first one you go through to get to the missiles). It only happened once or twice and went away when leaving and coming back.

[darcagn]

stuff

I don't think Tim's mistake was not testing it enough. I mean, I definitely felt that way when the second batch was released and a massive proportion of games were buggy--that was clearly not testing well enough, IMO. But in general, 6 firmware revisions across several months shows us that we can't expect entirely bug-free operation out of the gate from a product of this nature. We can demand that Tim fix it, but we don't need to, as he's been releasing updates in a timely manner. I think the actual mistake was not building in an update mechanism in the board. Sure, it would have increased costs, and hindsight is 20/20, but when you add in the cost of support and shipping blasters all over the world, it probably would have been worth it, plus it could have taken advantage of more features (like updatable/customizable palettes... I personally don't care about that feature but some people do).

[lettuce]

The addition to the through hole where the Famicom audio channels get mixed in is super nice. I was able to use a connector for that as well because of it. I can now remove kits from the AV Famicom without having to desolder anything. Works super nice if you have to test stuff a lot since it allows you to quickly disconnect it.

So is there a pad now to solder the expanded auido mod?

Also anyone know of places in the UK that sell the above pictured pin connection and sockets for quick release, what are they called/model type?

[darcagn]

So is there a pad now to solder the expanded auido mod?

It's a hole where you're expected to put a resistor and wire for your expansion audio pin.

Also anyone know of places in the UK that sell the above pictured pin connection and sockets for quick release, what are they called/model type?

Not in the UK, but for reference, these are what I used:

here and here

[lettuce]

Not in the UK, but for reference, these are what I used:

here and here

these seem to be a good match?....

http://www.ebay.co.uk/itm/PIN-HEADER-8- ... 3a84be56c1

http://www.ebay.co.uk/itm/40pcs-Dupont- ... 417971f247

[darcagn]

Yup, those should work great.

[ms06fz]

It took me a good 2-3 hours to get my PPU desoldered but, I did it! And apparently without causing any major damage to the PPU or the motherboard. So I've NESRGB-modded my NES, at least far enough to be able to test it out. I still need to wire up the new video connectors and deal with the voltage regulator (just running on the stock reg. for now...)

I guess this is another reason why it's good that I bought two NESRGB kits - I get a practice run on the NES before modding my Famicom.