schattenmann

So please now return to the original subject: difference in image processing between different SoC's. Whether the different "picture" is actually a quality improvement, or only perceived as being better quality, is interesting. One thing is sure, all of the image processing in a SoC takes places at the hardware level, so no driver can ever change that, contrary what keeps being suggested here.

That's not correct. In my first post I've tried to summarize what I had read here so far and I also was asking question / thinking loudly and in the end it was about hardware as well:

[...], so that everything becomes a matter of the hardware:

So why STB producers cannot modify the source codes to get a better picture quality ? Simply unskilled developers are the reason ?

Scaler = hardware. There is no software involved.[...]

which makes it hopeless to expect, that Broadcom boxes will ever have a "better" picture quality? Would the only solution be to use another chip? Is there a better one from Broadcom?

But everyone can read what you had answered to this... This dicussion doesn't have a chance to lead to something useful though, so I gave up on this.

[...]
If you had read the whole thread, you would have known that I had explained the whole issue already two times, and that this is a very short summary, because I don't want to explain it a THIRD time.

Funny you are using the term upscaling by the way. If your tv has 1920 by 1080 pixels and the original material is SD, it has to be scaled somewhere. If not, you'd have a very small picture... So it's scaled by the broadcaster, in the stb or in the television. What flavour exactly do you call "upscaling" and why would one be better than another, exactly?

DVBAPI drivers are nor closed as far as I understand it and so it should be possible for developers to improve them etc.

You definitely got that wrong. All drivers for broadcom SoC are closed source. And once again, even if the source was available, there would be nothing to "improve" because the driver (= software) is not in between the signal path.

No, it's "funny" how ignorant you are, simply assuming people don't read / understand what you and others write here, while you yourself seem not to understand (or not to want to understand) what people are saying. Be sure that I've read your replies with interest, but as I said: because we are talking about altering the signal, all the details about not altering it and stuff like SD = SD etc. are not of interest in this case.

I indeed didn't know if the drivers for broadcom SoC are closed source or not and that's was one of the questions. It's answered now and shows, that not much can be done, but your statement

And once again, even if the source was available, there would be nothing to "improve" because the driver (= software) is not in between the signal path.

is pretty much the opposite to this in my opinion or I'm totaly misunderstand what WanWizard says:

WanWizard, on 24 May 2012 - 17:11, said:

[...]Now, it could be that the Broadcoms are not as good as some other SoC's in postprocessing of the digital signal, or that the drivers of a particular platform don't allow you to tune the SoC properly, or that you have chosen the wrong setup or configuration. And that might result in a difference in the end result.

The only conclusion you can draw from this is that maybe one SoC has a better scaler/deinterlacers/whatever feature then another.

WanWizard, on 24 May 2012 - 17:11, said:
[...]Any feature you will enable will alter the signal, and while you might percieve it as being a better picture, technically it is altered from the original. Which perhaps means that the original input signal wasn't too good, so the SoC could visually improve something. Fine. [...]

Hello,

thanks for the detailed reply, WanWizard!

[...]
Any feature you will enable will alter the signal, and while you might percieve it as being a better picture, technically it is altered from the original. Which perhaps means that the original input signal wasn't too good, so the SoC could visually improve something. Fine. But that has nothing to do with picture quality, but with your personal perception of what is better...

While the term "picture quality" might be wrong technically based on your explanation, let's be honest: people just can't call it differently, since that's what it comes down to, if everyone is trying to value what (s)he sees. Since there are really a lot of statements on dozens of different forums, which state that "picture quality" of STI / sh4 boxes is (much) better compared to broadcom boxes, there simply *is* something common regarding this among very many people. So I think we shouldn't try to rip apart any terms used, but try to find common ground about the core issue.

It is resulting in this in my opinion

[...]Now, it could be that the Broadcoms are not as good as some other SoC's in postprocessing of the digital signal, or that the drivers of a particular platform don't allow you to tune the SoC properly, or that you have chosen the wrong setup or configuration. And that might result in a difference in the end result.

The only conclusion you can draw from this is that maybe one SoC has a better scaler/deinterlacers/whatever feature then another.

and in this, again:

[...]Any feature you will enable will alter the signal, and while you might percieve it as being a better picture, technically it is altered from the original. Which perhaps means that the original input signal wasn't too good, so the SoC could visually improve something. Fine. [...]

That's exactly what the whole thread is about as far as I understand it. At least that was what I've mentioned in my reply in post #54

Very many people are of the opinion, that SD material upscaled to 1080i "looks better" than the original. Not everyone probably, but very many, that's a fact. So we are already talking about upscaling, which seems to be handled better by STi boxes than by broadcom boxes.

Once malakudi has explained the driver situation (see both quotes in my reply in post #54 it became clear, that not much can be done for E2 being used on STi boxes. Ok and fine.

That's why I've asked about the apparently worse results of upscaling with E2 on *broadcom* boxes. Here the drivers problem doesn't matter, since there is no need to use the closed STAPI drivers. DVBAPI drivers are nor closed as far as I understand it and so it should be possible for developers to improve them etc.

Is this not possible?
If not: why? Would it only work with another chip (like STi)?

That's all I wanted to know, but replies like

Sigh... There is nothing wrong with the broadcom chipset! DIGITAL = DIGITAL = DIGITAL! SD = SD = SD!

are not very helpful and a bit one sided, sorry.

It's clear, that it's difficult to not possible to measure the "improvements" on the drivers, but see it from the other side: there are only voices about worse "picture quality" from the owners of broadcom / ALI boxes. STi people don't complain about. Often people were comparing the results directly at the same time and results are very often in favor of STi boxes, so there must be something technical responsive for this and that's what the thread is about to find out what it is and how it can be improved for broadcom boxes.

Under these circumstances I would say it doesn't matter, if "picture quality" is the correct term technically and also the whole stuff about not upscaled material is more or less pointless (although interesting to know), because it's about upscaling SD material, which is mostly considered better by many people, i.e. better than unchanged SD material and about the differences in doing so between STi and broadcom.