A500 rev.5 motherboards seems to have better picture quality VS Rev.6A

[turrican9]

So I have like 4 Amiga 500s with Rev. 6A motherboards and 3 with Rev.5 motherboards.

The first time I got hold of a Rev.5 board a few years ago I noticed it seemed to have noticeable better RGB picture quality VS 6A. Didn't think much of it back then but lately I have been messing around alot with various of my Amigas and noticed it again. I am using BenQ BL 702A and BL912 monitors connected to RGB. Pixel clock is finetuned for best picture of course.

Rev. 6A seems to have faint vertical bars and looks quite good. Rev.5 on the other hand has no faint vertical bars and looks even more vibrant with extremely sharp fonts. I could also see it on the blue whdload info window that comes up when loading whdload games. I tested several games and could see a distinct difference. It was extremely sharp and vibrant on Rev.5. Not that it looks bad on 6A but Rev.5 is almost emulator quality I would say. Extremely good picture quality on Rev.5 I have the same results on my other Rev.5s and 6As. 4 of these boards have been recapped. Not that it made any difference anyway because the old caps was still fine.

6A RGB seems to have very similar quality to A600 RGB. And the 1200 has the worst RGB quality in my opinion. Rev.5 A500 boards just has like extremely good quality. At least on my above mentioned screens.

Because of this I have decided to make one of my Rev.5 boards as my main A500 for ACA 500 + ACA 1221. Throw in an 8372A Agnus to get the 1MB chip on it.

As I said, I first noticed this when I got my first Rev.5 board a few years ago. Started loading a few floppy games and stuff and noticed it seemed to have extremely good picture quality. Noticeable better than my 6As. And again, not that the 6As have bad quality, just that the Rev.5s seems to have extremely good quality.

It kinda makes sense because Commodore was known for cost cutting strategies. For each revision they probably tried to save money where they could to make more money. There are probably some cheaper components in the RGB circuitry on the Rev. 6A VS Rev. 5. Maybe it would not be noticeable on an old CRT, but I see a clear difference on my BenQ screens. As I said, the Rev.5 is almost up to emulator quality.

Anyway, I thought this was interesting and wanted to mention it. Has anyone else noticed this difference?

Edit: I took the 8372A from one of my Rev. 6A boards, taped over pin 41 and put it into my Rev. 5 board. Reason I had to tape over pin 41 is to switch it into PAL. This needs to be done when you put one of these chips into a Rev.5 board, or it will run NTSC. I forgot this at first, and sure enough, it ran NTSC. Even the ACA 500 could not switch it to PAL. After I taped the pin, it ran in PAL and the ACA 500 did it's job in turning the trapdoor mem into chip ram. Just like it would do on a Rev. 6A. So no need to do the chip RAM mod as long as I have the ACA 500. I had some serious problems extracting the 8271 Agnus that was in this board. And yes, I have the correct extractor tool. Chip had probably never been out of the socket before. So it sat there for close to 32 years. Anyway, I ended up scratching the chip and bent a leg. Chip is fine though. I then proceeded to put the 8271 into the Rev. 6A board. This requires JP4 to be closed for it to work in a Rev. 6A board. So I added a solder blob there. So this Rev. 6A Amiga will work normally for floppy games and stuff. But not good for 1MB chip of course.

Been messing around in classic workbench and games with this Rev.5 board now and am more and more impressed by the picture quality. It's just extremely crisp and vibrant. Later I plugged in one of my 6A boards again and could see vertical bars and just not the same level of picture quality as the Rev. 5. Safe to say, I will transfer the RED/Green LED keyboard controller card and the red caps lock LED to one of my finest A500 keyboards, and also transfer this Rev. 5 motherboard to one of my finest A500 cases and make this my main ACA 500 + ACA 1221 Amiga 500. Or main Amiga 500 in other words.

[robinsonb5]

It would be interesting to know whether this is down to Denise revisions, the video hybrid or something else. (I'm guessing you have revision -01 hybrid in the Rev 5 boards, and -03 in the 6As?)

It would be especially interesting to know what happens if you swap hybrids - or even switch a 6A's hybrid for a modern replacement.

[dreadnought]

It's interesting but I'm always skeptical of such claims in regard to visuals because there are usually so many factors involved and the whole subject can be quite relative too. You'd also need to try it on a bigger sample to come up with some definitive conclusions (or have others to confirm this).

I'm also not sure what "up to emulator quality" means in this context.

[turrican9]

Originally Posted by robinsonb5

It would be interesting to know whether this is down to Denise revisions, the video hybrid or something else. (I'm guessing you have revision -01 hybrid in the Rev 5 boards, and -03 in the 6As?)

It would be especially interesting to know what happens if you swap hybrids - or even switch a 6A's hybrid for a modern replacement.

I have not tried to swap OCS Denise to ECS. No clue as to what revisions of the Hybrids are in there. Have to check it.

All my 6As shows almost the same image quality. Two of them is the later golden PCB 6As and two are the older green PCB 6As. My Rev. 5s shows the same quality between them. Which is quite a bit sharper and better VS my Rev 6As.

[turrican9]

Originally Posted by dreadnought

It's interesting but I'm always skeptical of such claims in regard to visuals because there are usually so many factors involved and the whole subject can be quite relative too. You'd also need to try it on a bigger sample to come up with some definitive conclusions (or have others to confirm this).

I'm also not sure what "up to emulator quality" means in this context.

I wouldn't call this subjective. My 3 Rev 5s simply shows a sharper and higher quality picture VS all my 6As. Maybe it's because the 6As produce faint vertical lines when fintuned on the above mentioned screens and Rev 5s don't.

I know it's not a huge sample. And I have only tested on the above mentioned screens. That's why I asked if anyone else has seen any difference. Like I said, 3 Rev 5s and 4 Rev 6As. Where two of the 6As are the older green PCB and two are the later golden colour PCBs. Difference is big enough for me to use Rev. 5 as main A500.

By emulator quality I mean it would be almost as sharp as if you ran an Amiga 500 emulator on a modern PC screen without stretching the picture. More like a figure of speech to try and say it's extremely good image quality.

As I said, I am interested if other people here have both Rev 5s and 6As and can spot a difference.

[dreadnought]

Originally Posted by turrican9

By emulator quality I mean it would be almost as sharp as if you ran an Amiga 500 emulator on a modern PC screen without stretching the picture. More like a figure of speech to try and say it's extremely good image quality.

I thought that's what you meant and one of the reasons I mentioned that the whole topic is subjective. That's because for me "emulator quality" is actually inferior, at least when it comes to games. They were designed for, and on CRTs, and look awful when displayed with supersharp square pixels on modern monitors.

Sometimes also a person can auto-suggest him or herself, especially when comparing things on the same display. It happened to me many times (I have an OCD and test a lot of solutions).

That said, if you tested it on 7 boards and the results are consistent it's not actually that a small sample. It'd be interesting to hear a technical explanation for that whole thing. Seeing some photos could also help (though I realise it's hard to capture minuscule differences on live displays).

I have a 6A but don't see any vertical lines, perhaps because I use a CRT TV. Could be similar to a non-Lumafixed c64, on which the noise is more pronounced on an LCD display than on a CRT.

[turrican9]

Originally Posted by dreadnought

I thought that's what you meant and one of the reasons I mentioned that the whole topic is subjective. That's because for me "emulator quality" is actually inferior, at least when it comes to games. They were designed for, and on CRTs, and look awful when displayed with supersharp square pixels on modern monitors.

Sometimes also a person can auto-suggest him or herself, especially when comparing things on the same display. It happened to me many times (I have an OCD and test a lot of solutions).

That said, if you tested it on 7 boards and the results are consistent it's not actually that a small sample. It'd be interesting to hear a technical explanation for that whole thing. Seeing some photos could also help (though I realise it's hard to capture minuscule differences on live displays).

I have a 6A but don't see any vertical lines, perhaps because I use a CRT TV. Could be similar to a non-Lumafixed c64, on which the noise is more pronounced on an LCD display than on a CRT.

It is indeed very subjective. But the fact of the matter is that the Rev.5 boards seems to probably have a cleaner/stronger RGB signal which gives it a 'superior' image VS Rev 6A tested on the aboove mentioned screens. So a technical better signal/image. Anyway, when using my particular LCDs which are 17 inch 5:4 I think Amiga games looks great. I cannot stand CRTs anymore. But that's my opinion. and when using LCDs like these I think the picture looks superior on the Rev.5 with no vertical stripes and slight blurr like the 6A has. And again, not that the 6A has bad quality, just that the Rev 5 looks superior. cleaner signal.

This is not auto suggesting. As mentioned earlier, I have always wanted to use the 6A boards. Especially for the ACA 500. And had noticed the image quality on these Rev 5s a few years ago but didn't really think about it until recently when I started using them again. The difference is big enough that I can instantly see it. As I said, extremely clear, sharp and vibrant picture which just pops at me. There is a noticeable difference.

This is not something new really. People, including me have seen the Amiga 1200 having a little different RGB picture VS A500s and A600s. Someone explained to me that it was because the A1200 has a different video DAC. In my opinion the A1200 has the weakest RGB signal out of the above mentioned. So makes sense that some parts of the RGB circuitry in A500 Rev.5 is different VS Rev 6A. If this difference would show on a CRT I don't know, but to me it seems like the Rev 5 has a cleaner/stronger RGB signal.

[turrican9]

And I am by no means a Rev. 5 fanboy. On the contrary I was a Rev.6 fanboy. Because my original Amiga 500 from 1991, which I still have is a Rev.6A with the golden colour PCB. But when I see the Rev.5 has this great picture quality it made me put an 8372A Fat Agnus in it to get 1MB chip and to use it with my ACA 500 as my main Amiga 500.

That should say something about how much I notice the picture quality difference. Doing this even though I was a real 6A fan. Nostalgia is strong. One often like best what one used back in the day.

[Damion]

I've noticed here on a CRT at least, that the A1200 has obviously worse RGB quality than Rev. 6 A500 and A2000. (The A1200 exhibits more "ghosting" to the right of text characters, which looks partially like an impedance mismatch somewhere... could be the case since I made the BNC cable myself.) I only have one functioning Rev. 5 A500 (at least it *was* functioning), I'll have to pull it out and check.

[turrican9]

Originally Posted by Damion

I've noticed here on a CRT at least, that the A1200 has obviously worse RGB quality than Rev. 6 A500 and A2000. (The A1200 exhibits more "ghosting" to the right of text characters, which looks partially like an impedance mismatch somewhere... could be the case since I made the BNC cable myself.) I only have one functioning Rev. 5 A500 (at least it *was* functioning), I'll have to pull it out and check.

I have the Indivision AGA MK2 on my A1200. Would never use it with RGB. Looks far worse than my other Amiga models. With the MK2 it looks good of course.

[turrican9]

So I wanted to try and switch denise in my Rev 5 board to see if that was part of why it has superior image quality but Rev. 6A and Rev. 5 uses the same 8362R8 Denise. I remembered wrongly and thought the Rev. 6A had a later ECS Denise. It's the A500 Plus which has the Super denise

So Denise has nothing to do with the difference. Was messing around with my A600s today. They have about the same picture quality as my A500 6As. Good picture but faint vertical bars and a very slight blur. Again, Rev. 5, no visible vertical bars and extremely sharp and vibrant picture.

And again, this is on my BenQ BL 702As and BL 912. But even still, this is an indication that something is different in the RGB output circuitry on the Rev 5s. Giving it higher quality picture. I would still like other people to test this and give feedback.

Edit: Just as a note. All my testing is in PAL mode. Have not tried NTSC. But I would assume it would have superior image quality in NTSC aswell

[Mathesar]

I have a rev 5 and severe vertical lines that annoy me a lot.
I have often wondered what the cause of these lines is. When you look at the Amiga 500 schematics you'll see that the video signal from Denise is also passed through 2 245 buffers and the "vidiot" circuit. My guess was always that the vertical lines are either caused by glitches from Denise itself or power supply crosstalk through the buffers and/or the vidiot circuit.

It is that I don't have much spare time otherwise I would probably try to design a new, modern vidiot which includes some proper power supply filtering and a video reclocker to get rid of any glitches.

[robinsonb5]

Originally Posted by Mathesar

I have often wondered what the cause of these lines is.

I'm pretty sure these lines are an interference pattern between the number of pixels the monitor is expecting to see in a line, and the number of pixels the Amiga actually produces. That's why changing the monitor's pixel clock setting affects the lines (and if you're really lucky, can get rid of them completely.)

I don't know for sure, but I suspect the problem is high-frequency ripple coming from the 7Mhz clock or one of its subdivisions.

With an analogue monitor it doesn't matter because the ripple's in sync with the pixel clock (it may even *be* the pixel clock) so any artifacts are consistent from pixel to pixel - the average level over the duration of the pixel is correct and it all ultimately gets smoothed out by the CRT and its phospors.

A digital monitor, on the other hand, samples the incoming signal at regular intervals. If its pixel clock doesn't perfectly match the Amiga's pixel clock (and it probably doesn't, because it's expecting to see a PC screenmode) then it will sample every pixel at a different point of the pixel's duration: for one pixel it'll be sampled near the end of its duration when everything's settled down; ten pixels later the monitor might be sampling at a noisy part of the pixel's duration, and see a level slightly lighter or darker than it should. As the clock drifts back and forward between stable and ripply parts of each pixel's duration we see vertical bars on the monitor.

So, again, I'm curious to know which revision vidiot is fitted to the A500s with "good" and "bad" picture quality. Or whether any of the A500s or their PSUs have had replacement capacitors.