Why were most crt TVs classified as 4:3 when they were 3:2?

[Einzelherz]

The best method is to run the test suite's grid on as many consoles as you can through your set to determine where the overscan lives. And after accepting the fact that not all games on all consoles run at their full res. And also accepting that the timing from each system varies, e.g. my Genesis is shifted to the right compared to my SNES.

[mortron]

I've learned that a good TV/Monitor like a good speaker and is sensitive to what it's fed. Einzelherz is right, not all systems are the same. The SNES Test Suite actually has 2 different grids, because not all games output at the same resolution (I guess?).

I have never heard of mention of 3:2 resolution in my time working in stores that sold televisions. Then again, I just found out there are 16:10 monitors, so anything is possible

[gray117]

I've learned that a good TV/Monitor like a good speaker and is sensitive to what it's fed. Einzelherz is right, not all systems are the same. The SNES Test Suite actually has 2 different grids, because not all games output at the same resolution (I guess?).

I have never heard of mention of 3:2 resolution in my time working in stores that sold televisions. Then again, I just found out there are 16:10 monitors, so anything is possible

No, the whole topic is moronic...You will find the geometry of a crt will vary on any quality of screen, sometimes quite dramatically, even across identical resolutions depending on the signal particulars. For most people with tv you just lived with it, generally with a fair amount of compensating overscan, and usage was often limited to 1 or 2 devices. With monitors, which generally saw a greater range of signal inputs, geometry controls were a must.

The 4:3 standard was based on physical measurement. 16:10 and 16:9 were two competing widescreen standards born in a confusing era of digital conversion, where it was still mostly centered around physical dimensions but of course the fidelity/pixel dimensions was also becoming increasingly important.


**Edit: sorry didn't actually know the snes that well. Turns out nintendo themselves may have actually assumed that ideally you'll have black borders on left and right. Either that, or they didn't care - this was an era when any game image on a tv was still a bit of miracle... Some third party developers on snes would actually author content assuming there was no black border... The point kind of stands to the fact that both made assumptions about how it would be displayed in 4:3 and the kind of mess we all just put up with In most other environments one would assume image would be filling 4:3 screen**
Taking the snes as an example most snes games are generally 256x224 (some 240) ... regardless of pixel aspect ratio, exact resolution, or number of vertical lines, it was generally accepted the idea was this should fill a full 4:3 screen. Clearly in terms of squares neither of these resolutions == 4:3 ... but they designed the games around the idea this would be represented in 4:3.

Pixel aspect ratios are a fucking mess, but were relied upon to (hopefully) stretch/squash authored content to hopefully fill a type screen in a correct manner - but of course there was no guarantee the end user would have set everything up correctly, and plenty of content/equipment providers who had their own opinion on the best way to do something.

It was not until HD (ready) [and the general availability of more capable processors] that the standard came to be expressed in (square) pixel resolution, and 'helped' simplify the widescreen manufacturing choice to largely center around 16:9, near guaranteed support for PAL 60hz, and helped to really nail content authoring standards from the legacy of competing analogue standards/supporting frameworks/opinions.

[Sumez]

I think this article does a good job at summing up the issue of pixel resolution vs. physical aspect ratio:

http://www.gamasutra.com/blogs/FelipePe ... _ratio.php