Jump to content


Photo

Since PLi 2.1 - A/V-Settings - Scaler Sharpness wrong target parameter

ET9000

  • Please log in to reply
38 replies to this topic

Re: Since PLi 2.1 - A/V-Settings - Scaler Sharpness wrong target parameter #21 Qu@rk

  • Senior Member
  • 156 posts

+2
Neutral

Posted 19 August 2011 - 13:46

I understand your concern. I have more than 20 years prof. home cinema experience, ISF calibrator and I am an Electronics and IT Engineer.
Maybe this helps. ;)

The difference may be subtle if you compare at 720p. At SD it is extremely good visible.
I have tested this with AutoResolution always at the native resolution where this setting has the biggest effect.

But during normal, long time viewing, the more natural picture is a huge difference (not only for me, but also for some "videophile" users).

Re: Since PLi 2.1 - A/V-Settings - Scaler Sharpness wrong target parameter #22 Qu@rk

  • Senior Member
  • 156 posts

+2
Neutral

Posted 23 August 2011 - 10:25

I have got the confirmation also from some VU and DM 8000 users.
Setting scaler sharpness (and sharpness) to 0 is the best setting.
This effect is well known and has been discussed very often.

Re: Since PLi 2.1 - A/V-Settings - Scaler Sharpness wrong target parameter #23 Qu@rk

  • Senior Member
  • 156 posts

+2
Neutral

Posted 23 August 2011 - 15:08

Please let me know:
Did you change something in E2 regarding this topic?

Beta tester and moderator of the CT forum "z-pg-3" wrote something about a change. He wrote that the mid position in the AV-settings is "0" now?

This would be a very inconsistent und illogical solution.
Please, are you able to confirm or deny this?

The solution with an ON/OFF switch would have been so elegant and user friendly.

Please let me know. Thanks.

Re: Since PLi 2.1 - A/V-Settings - Scaler Sharpness wrong target parameter #24 pieterg

  • PLi® Core member
  • 32,766 posts

+245
Excellent

Posted 23 August 2011 - 20:55

I have got the confirmation also from some VU and DM 8000 users.
Setting scaler sharpness (and sharpness) to 0 is the best setting.
This effect is well known and has been discussed very often.


Wouldn't know about VU, but I've just tested this on my 8k.
And there 0 is the least sharp setting, and somewhere around 13 is it neutral.
(though it is hard to give an exact judgment on differences between -say- 12,13,14)

So we cannot create an 'off' setting, that won't give the desired effect on all hardware.

Re: Since PLi 2.1 - A/V-Settings - Scaler Sharpness wrong target parameter #25 Erik Slagter

  • PLi® Core member
  • 46,960 posts

+541
Excellent

Posted 23 August 2011 - 21:23

Wouldn't know about VU, but I've just tested this on my 8k.
And there 0 is the least sharp setting, and somewhere around 13 is it neutral.
(though it is hard to give an exact judgment on differences between -say- 12,13,14)


In my experience, on the dm8000, 0 for scaler sharpness is really "off". It
shows sd quality in all of it's ugly appearance ;-) Therefore I always have
it at ~50%. Of course you shouldn't compare to upscaled hd channels as
they're always sharpened already.

* Wavefrontier T90 with 28E/23E/19E/13E via SCR switches 2 x 2 x 6 user bands
I don't read PM -> if you have something to ask or to report, do it in the forum so others can benefit. I don't take freelance jobs.
Ik lees geen PM -> als je iets te vragen of te melden hebt, doe het op het forum, zodat anderen er ook wat aan hebben.


Re: Since PLi 2.1 - A/V-Settings - Scaler Sharpness wrong target parameter #26 pieterg

  • PLi® Core member
  • 32,766 posts

+245
Excellent

Posted 23 August 2011 - 21:58

In my experience, on the dm8000, 0 for scaler sharpness is really "off". It


weird, here 0 (left most) is really fuzzy, a lot of smoothing/smudging going on, definitely not 'off'.

Re: Since PLi 2.1 - A/V-Settings - Scaler Sharpness wrong target parameter #27 Qu@rk

  • Senior Member
  • 156 posts

+2
Neutral

Posted 24 August 2011 - 08:30

I fear that I have done a big mistake, forcing a solution for ET 9000 here.

Please, do not change the current behavior (0 = off) until you have found a logical and user friendly solution.

The current behavior is:
0 = OFF
1 ... 26 = from smooth to sharp (very ugly in my opinion)

It would make absolute no sense to insert 0 as a mid position!

In the meantime, I have so many positive feedback in the Clarke Tech ET9000 forum.
Even the users who questioned thjs at the first time, had to admit, that the picture quality is improved a lot with this setting on the ET9000 after clean testing.

Please leave the users the choice, whether to turn off picture enhancements like scaler sharpness etc.

That's the big difference of E2 receivers, the flexibility.

If you are not sure, please leave it the way it is on the ET9000 (0 = off, left most, 1 ... 26).

I can not speak for VU or DM receivers.


Re: Since PLi 2.1 - A/V-Settings - Scaler Sharpness wrong target parameter #28 adri

  • Senior Member
  • 372 posts

+5
Neutral

Posted 24 August 2011 - 08:34


In my experience, on the dm8000, 0 for scaler sharpness is really "off". It


weird, here 0 (left most) is really fuzzy, a lot of smoothing/smudging going on, definitely not 'off'.

I have the same experience on my DM8000.
Setting the scaler sharpness to 0, creates a really fuzzy and ugly picture on SD channels.
I prefer the setting around 14-16 for SD channels.

Adri.

Re: Since PLi 2.1 - A/V-Settings - Scaler Sharpness wrong target parameter #29 Qu@rk

  • Senior Member
  • 156 posts

+2
Neutral

Posted 24 August 2011 - 09:56

Please do not misunderstand the effect. The scaler sharpness 13 smoothes the picture with SD channels. So bad SD channels do look a bit better than turing off scaler sharpness with 0. Without the smoothing effect of scaler sharpness 13 at SD channels you see all the artefacts of bad (low bit rate) transmissions.

This effect is even much stronger if you have activated agressive video enhancement in your TV (like it is standard in many TVs).

We can not discuss what looks better or worse for some people, that's a matter of taste.
Everybody can choose the best setting for his taste.

But we have to choose a technical correct and logical user interface, so the user can decide what's best for him.

I would have no problem to set a scaler sharpness on/off switch to ON and a default value of 13 as the standard installation setup.
If the community prefers this, no problem.

But the more sophisticated user should have the choice to turn off the video enhancment scaler sharpness, e.g. via an ON/OFF switch which results in "0" at scaler sharpness.


Re: Since PLi 2.1 - A/V-Settings - Scaler Sharpness wrong target parameter #30 Erik Slagter

  • PLi® Core member
  • 46,960 posts

+541
Excellent

Posted 24 August 2011 - 11:38

weird, here 0 (left most) is really fuzzy, a lot of smoothing/smudging going on, definitely not 'off'.


Setting "0" is imho comparable to the picture when the drivers did not
yet support this setting.

* Wavefrontier T90 with 28E/23E/19E/13E via SCR switches 2 x 2 x 6 user bands
I don't read PM -> if you have something to ask or to report, do it in the forum so others can benefit. I don't take freelance jobs.
Ik lees geen PM -> als je iets te vragen of te melden hebt, doe het op het forum, zodat anderen er ook wat aan hebben.


Re: Since PLi 2.1 - A/V-Settings - Scaler Sharpness wrong target parameter #31 Erik Slagter

  • PLi® Core member
  • 46,960 posts

+541
Excellent

Posted 24 August 2011 - 11:43

I have the same experience on my DM8000.
Setting the scaler sharpness to 0, creates a really fuzzy and ugly picture on SD channels.
I prefer the setting around 14-16 for SD channels.


What you prefer may not actually be what's "neutral" ;-) Please accept
that scaled SD is simply ugly and fuzzy and that some (extra) sharpening
is necessary. That doesn't mean the extra sharpening is "neutral" so it
shouldn't be the "neutral" value. The neutral value should mean that no
processing is done. That way you can start tweaking from a completely
neutral basis, with little chance of different settings influencing each
other.

* Wavefrontier T90 with 28E/23E/19E/13E via SCR switches 2 x 2 x 6 user bands
I don't read PM -> if you have something to ask or to report, do it in the forum so others can benefit. I don't take freelance jobs.
Ik lees geen PM -> als je iets te vragen of te melden hebt, doe het op het forum, zodat anderen er ook wat aan hebben.


Re: Since PLi 2.1 - A/V-Settings - Scaler Sharpness wrong target parameter #32 pieterg

  • PLi® Core member
  • 32,766 posts

+245
Excellent

Posted 24 August 2011 - 11:52

weird, here 0 (left most) is really fuzzy, a lot of smoothing/smudging going on, definitely not 'off'.


Setting "0" is imho comparable to the picture when the drivers did not
yet support this setting.


ok, let's put it another way, do you see a big difference between 0 and 1, or do you see only a tiny difference.
Or in other words, is 0 more similar to 13, or more similar to 1?

In my opinion, 0..26 should probably have been more or less linear. On my 8k this is the case.
And the center position (value 13, which is also the driver default when you boot without starting e2) in my opinion corresponds to what things looked like before the scaler sharpness setting was added.

On the et9k, where the scaler sharpness was added in the more recent history, I know that for sure (by going back to a driver version before the scaler sharpness setting was introduced)
What is currently 0, corresponds to what things looked like before the scaler sharpness was added.
And, this 0 corresponds to what my 8k produces with its scaler sharpness set to 13. And 8k on 0 roughly corresponds to et9k on 1.

Re: Since PLi 2.1 - A/V-Settings - Scaler Sharpness wrong target parameter #33 Erik Slagter

  • PLi® Core member
  • 46,960 posts

+541
Excellent

Posted 24 August 2011 - 12:03

In my opinion, 0..26 should probably have been more or less linear.
On my 8k this is the case.


Seconded.

And the center position (value 13, which
is also the driver default when you boot without starting e2) in my
opinion corresponds to what things looked like before the scaler
sharpness setting was added.


And that is not my experience. But I must admit I've playing with that
setting some time ago, I might have been changed in the meantime.

Okay, so I need to do some playing with this setting again.

* Wavefrontier T90 with 28E/23E/19E/13E via SCR switches 2 x 2 x 6 user bands
I don't read PM -> if you have something to ask or to report, do it in the forum so others can benefit. I don't take freelance jobs.
Ik lees geen PM -> als je iets te vragen of te melden hebt, doe het op het forum, zodat anderen er ook wat aan hebben.


Re: Since PLi 2.1 - A/V-Settings - Scaler Sharpness wrong target parameter #34 adri

  • Senior Member
  • 372 posts

+5
Neutral

Posted 24 August 2011 - 20:24

I have the same experience on my DM8000.
Setting the scaler sharpness to 0, creates a really fuzzy and ugly picture on SD channels.
I prefer the setting around 14-16 for SD channels.


What you prefer may not actually be what's "neutral" ;-) Please accept
that scaled SD is simply ugly and fuzzy and that some (extra) sharpening
is necessary. That doesn't mean the extra sharpening is "neutral" so it
shouldn't be the "neutral" value. The neutral value should mean that no
processing is done. That way you can start tweaking from a completely
neutral basis, with little chance of different settings influencing each
other.

Erik,

I have checked the scaler sharpness and compared it against a recording on a DVD of the same program.
Using 0 or anything below 10 on my DM8000, results in a picture, which is much softer, fuzzier and uglier, than when playing the same program as a recording from the DVD.
No scaler sharpness should correspond to a similar picture as the DVD I think.

Adri.

Re: Since PLi 2.1 - A/V-Settings - Scaler Sharpness wrong target parameter #35 Erik Slagter

  • PLi® Core member
  • 46,960 posts

+541
Excellent

Posted 25 August 2011 - 10:33

I have checked the scaler sharpness and compared it against a recording on a DVD of the same program.
Using 0 or anything below 10 on my DM8000, results in a picture, which is much softer, fuzzier and uglier, than when playing the same program as a recording from the DVD.
No scaler sharpness should correspond to a similar picture as the DVD I think.


Okay, so I am wrong here ;-)

* Wavefrontier T90 with 28E/23E/19E/13E via SCR switches 2 x 2 x 6 user bands
I don't read PM -> if you have something to ask or to report, do it in the forum so others can benefit. I don't take freelance jobs.
Ik lees geen PM -> als je iets te vragen of te melden hebt, doe het op het forum, zodat anderen er ook wat aan hebben.


Re: Since PLi 2.1 - A/V-Settings - Scaler Sharpness wrong target parameter #36 Qu@rk

  • Senior Member
  • 156 posts

+2
Neutral

Posted 26 August 2011 - 08:50

The Clarke Tech team released beta drivers which should have the same behaviour as the DM8000.

The mid position (13) for scaler sharpness in the new drivers equates to "0" of the previous (current) drivers.
Below 13 the picture will be softer, above 13 the picture will be sharper.

I do not know when these drivers will be on the update feed. I am sure you (PLi team) know that. ;)

Thanks for your support.



Re: Since PLi 2.1 - A/V-Settings - Scaler Sharpness wrong target parameter #37 littlesat

  • PLi® Core member
  • 56,508 posts

+693
Excellent

Posted 26 August 2011 - 08:51

Where can we find those beta drivers?

Re: Since PLi 2.1 - A/V-Settings - Scaler Sharpness wrong target parameter #38 Qu@rk

  • Senior Member
  • 156 posts

+2
Neutral

Posted 26 August 2011 - 11:46

Clarke Tech Forum: Neue Treiber 25.08.2011 als Test von Walli bereitgestellt!

http://www.clarke-su...ull=1#post29992

Re: Since PLi 2.1 - A/V-Settings - Scaler Sharpness wrong target parameter #39 Erik Slagter

  • PLi® Core member
  • 46,960 posts

+541
Excellent

Posted 26 August 2011 - 17:07

>> I have checked the scaler sharpness and compared it against a recording on a DVD of the same program.
>> Using 0 or anything below 10 on my DM8000, results in a picture, which is much softer, fuzzier and uglier, than when playing the same program as a recording from the DVD.
>> No scaler sharpness should correspond to a similar picture as the DVD I think.

Okay, so I am wrong here ;-)


So I tried it yesterday.

Result: sharpness is a lineair function from 0 to 26. The step between 0
and 1 is equal to the step between 1 and 2. 0 looks like no additional
sharpening, 13 like medium sharpening and 26 like "too much".

This is exactly like I mentioned before. I do have all other sharpening on
the dreambox set to "off" and on the tv to "neutral", maybe that matters?

* Wavefrontier T90 with 28E/23E/19E/13E via SCR switches 2 x 2 x 6 user bands
I don't read PM -> if you have something to ask or to report, do it in the forum so others can benefit. I don't take freelance jobs.
Ik lees geen PM -> als je iets te vragen of te melden hebt, doe het op het forum, zodat anderen er ook wat aan hebben.




Also tagged with one or more of these keywords: ET9000

1 user(s) are reading this topic

0 members, 1 guests, 0 anonymous users