24" Widescreen CRT (FW900) From Ebay arrived,Comments.

Anyhow, I got out my macro lens to see what can actually be resolved at various resolutions. This is from a Dell P1130:

View attachment 106081

So I guess we should stick with 1024x768!

The manual says it recommends 1280 × 1024. You can see the pattern decently at that resolution, actually. It's kind of still there at 1600x1200, but at 2496x1872, the pattern is definitely gone =p.
 
The manual says it recommends 1280 × 1024. You can see the pattern decently at that resolution, actually. It's kind of still there at 1600x1200, but at 2496x1872, the pattern is definitely gone =p.

Yeah, all you have to do is blur your vision a little bit and you can still see the pattern at 1600x1200. I mean, if you consider the pattern as 5 lines from left to right (3 color, with 2 black) then 1600x1200 still has headroom. It's just that beam isn't completely focused. It's only in the 1872p image that you see it's trying to fit a 5-line image into 4 lines. Information has definitely been lost

But very cool picture though! I should try this on my LaCie. Can you post the source file Ashun ?
 
Last edited:
Just got in that HDMI adapter CommandoSnake mentioned earlier, and it's pretty awful.

While it'll display an image at a 345 MHz pixel clock, the quality is terrible. In fact, anything past 181 MHz shows green lines and dots on the dark parts of the image. The auxiliary USB power made no difference. I'm not sure why we're getting such different results, but for now, even though it's cheap, don't bother with this one.

Are you outputting RGB or YCbCr? Try forcing YCbCr and that will fix the image quality issues that you are mentioning.
 
CommandoSnake Thank you! Although now I'm having a lot of trouble getting custom resolutions to work in YCbCr. The max that will show up in nvidia's control panel is 2048x1536 at 74 Hz (334.46 MHz). A single Hz higher and nothing shows up in the list. I've been messing around in CRU for about 40 minutes now with nothing to show for it.

Enhanced Interrogator It's hard to get those shots in focus and steady; the mechanical slap of the shutter blurs the images. I think my camera can do a "silent shutter" shutterless mode, so I'm looking into getting much better shots.
 
CommandoSnake Thank you! Although now I'm having a lot of trouble getting custom resolutions to work in YCbCr. The max that will show up in nvidia's control panel is 2048x1536 at 74 Hz (334.46 MHz). A single Hz higher and nothing shows up in the list. I've been messing around in CRU for about 40 minutes now with nothing to show for it.

This might be a similar issue to how Windows will ignore custom resoultions for displayport when they're above the max pixel clock on your display's EDID.

What's the max pixel clock on your monitors EDID?
 
Ashun have you tried Nvidia's custom resolution tool to get past that max resolution? Also, what's your monitors stated EDID in CRU (before modification). You click on "edit" next to your monitor name to see it
 
I mean, he's a bit of a dick, but he's also right. I had to spend a few minutes scanning the last few pages to find what adapter was being talked about.

That said, XoR, you really need to tone it down, because I do like having your perspective in the forum, but you're also pissing a lot of people off with your tone.
 
If you want to see the pixels clearly then something around 800x600 is the way to go (I use that for Windows 98 GUI where I dont surf the web). I would say upto 1280x1024 is reasonably sharp too as someone stated above but its not as good as 800x600.

The image also gets considerably more blurry with refresh rates higher than 60Hz, especially with higher resolutions. I think the lower refresh rates help with sharpness because there is a nice amount of strobing for my eyes too (? not sure how else to call it) and is possibly more in line with the persistence of the phosphor layer. Even though 60Hz flickers, its not that apparent in most games (so I use it for these games, unless its a competitive shooter where I prefer 160Hz on lower resolutions).

If you have a newer nVidia graphics card you can try one of my favorite modes:

- Using CRU add [email protected] resolution (those with Sunix or similar can even make that into 16:10 [email protected]). All are "automatic CRT standard" timings.
- Set nVidia DSR to 4x and 0% smoothing.
- Pick a game from this list that has SGSSAA written next to it, for example Call of Duty 4 Modern Warfare. I recommend this game or any other game that scales UI well in 4K and beyond, works with custom resolutions and has reasonably high resolution textures. Choose older games for graphics cards older than 1080ti which for example yields around I think 70FPS in COD4 using this method. The list: docs.google.com/spreadsheets/d/1ekUZsK2YXgd5XjjH1M7QkHIQgKO_i4bHCUdPeAd6OCo/pub?output=html#
- Install "nVidia Profile Inspector" (NPI) app: https://forums.guru3d.com/threads/nvidia-inspector-introduction-and-guide.403676/
- Open NPI, write the name of your game on top and find it. Change these settings and save them with "Apply changes" button:

Antialiasing compatibility = find the relevant flag for your game in the list: docs.google.com/spreadsheets/d/1ekUZsK2YXgd5XjjH1M7QkHIQgKO_i4bHCUdPeAd6OCo/pub?output=html#
Antialiasing - Mode = Override any application setting
Antialiasing - Setting = 2x (2x Multisampling) (you can use 4X or 8Q multisampling later if you will have some FPS headroom, but probably wont)
Antialiasing - Transparency Supersampling = 2x Sparse Grid Supersampling (you can use 4X or 8X SGSS later if you will have some FPS headroom, but probably wont)
Anisotropic filtering mode = User-defined / Off
Anisotropic filtering setting = 16x
Texture filtering - LOD Bias (DX) = -1.1250 (thats a negative number)
Texture filtering - LOD Bias (OGL) = -1.1250 (thats a negative number)
Texture filtering - Quality = High Quality

Here is a screenshot of the above settings already in place: https://i.imgur.com/wgEsJGL.png

- Start the game and change the in-game resolution to [email protected]
- Activate in-game VSync (do it even if you hate the input lag!)
- Using CRT OSD controls, make the image as small as possible while maintaining the aspect ratio, 90 contrast and dimly lit (not fully dark) room. Change CRT OSD Brightness so that full black image cannot be seen and adjust the in-game gamma slider if you see black crush during gameplay.
- Play the game The Way It's Meant to be Played™®

The result demonstrates that even the diminishing returns with fixed aperture grill can be seen - or rather felt because it is something you just need to see for yourself in person and is hard to explain or show on a pattern test.

The full combination of 60Hz, DSR, SGSSAA, VSync, highest possible resolution on your CRT and smallest CRT image produces a truly unique image that one needs to see to truly appreciate.

Let me know your thoughts on the result if you do try it, I know its a rather involved process (if you are not already doing this of course), but its worth the shot :)
 
Last edited:
I understand why he's frustrated, but a bit of grace and politeness goes a long way. His tone suggests he thinks he's entitled to this information (which is being freely provided by others), by birthright.

Instead of

"How dare you not include the link", try

"Hey, mind linking to that piece of information please, thanks, appreciate it"
 
If you want to see the pixels clearly then something around 800x600 is the way to go (I use that for Windows 98 GUI where I dont surf the web). I would say upto 1280x1024 is reasonably sharp too as someone stated above but its not as good as 800x600.

The image also gets considerably more blurry with refresh rates higher than 60Hz, especially with higher resolutions. I think the lower refresh rates help with sharpness because there is a nice amount of strobing for my eyes too (? not sure how else to call it) and is possibly more in line with the persistence of the phosphor layer. Even though 60Hz flickers, its not that apparent in most games (so I use it for these games, unless its a competitive shooter where I prefer 160Hz on lower resolutions).

I wasn't aiming to actually see the pixels clearly, I was aiming for all pixels are actually shown instead of hidden behind the shadow mask.

As for why 60 Hz is super sharp, I think I read somewhere it's because AC power is at 60 Hz and so any other refresh rate not a multiple of 60 Hz has some power interference. It could be wrong, but it made perfect sense to me. If this were true, then multiples of 50 Hz would be the sharpest in Europe.
 
Enhanced Interrogator I didn't even know that was there!

I'll get to the EDID, but I did some more testing. Custom resolutions added in CRU will only show up in the Nvidia control panel if they're below 340 MHz. So for instance, 2300x1737x60 shows up (399.99 MHz), but 2300x1740x60 does not (340.56 MHz). The Benfei can do 2300x1737x60 in YCbCr no problem. Following your suggestion, the Nvidia custom res tool can get higher frequencies to populate the list, but with these, Nvidia only allows RGB, not YCbCr, and the screen goes distorted and green like the image I posted earlier.

Perhaps if there were a way to get Nvidia custom resolutions to use YCbCr, I could get the Benfei a little higher, but only up to a certain point. It'll display an RGB res all the way out to 365 MHz, but at that point the screen starts flickering like crazy, and anything past won't display at all. That's likely the limit of the chip.

But back to the EDID, the Dell P1130 EDID lists the max pixel clock at 290 MHz. The Benfei adapter has an extension block that lists a max TMDS clock of 225 MHz, but those don't seem to matter because I can hit higher without any modification. The hdmi.dat listed on the first CRU page overrides the TMDS to 340 MHz, but changing that, or even enabling DVI dual-link operation has no effect.


And I'm actually a bit surprised that more people didn't know exactly which adapter I was talking about. When I saw CommandoSnake's post, I thought there'd be huge interest in an $8 adapter that could hit at least 316 MHz. When I first got it in and was getting green artifacts all over the screen because I didn't know about the YCbCr thing, I didn't want to link people to something that didn't work.
 
Anyhow, I got out my macro lens to see what can actually be resolved at various resolutions. This is from a Dell P1130:

View attachment 106081

So I guess we should stick with 1024x768!
Oh good man, thank you for this eye-candy :)


BTW. I sincerely apologize for my rudeness earlier. Hope you will forgive me. I will try to do better and not make posts before my morning coffee ever again :dead:
 
Enhanced Interrogator I didn't even know that was there!

I'll get to the EDID, but I did some more testing. Custom resolutions added in CRU will only show up in the Nvidia control panel if they're below 340 MHz. So for instance, 2300x1737x60 shows up (399.99 MHz), but 2300x1740x60 does not (340.56 MHz). The Benfei can do 2300x1737x60 in YCbCr no problem. Following your suggestion, the Nvidia custom res tool can get higher frequencies to populate the list, but with these, Nvidia only allows RGB, not YCbCr, and the screen goes distorted and green like the image I posted earlier.

Perhaps if there were a way to get Nvidia custom resolutions to use YCbCr, I could get the Benfei a little higher, but only up to a certain point. It'll display an RGB res all the way out to 365 MHz, but at that point the screen starts flickering like crazy, and anything past won't display at all. That's likely the limit of the chip.

But back to the EDID, the Dell P1130 EDID lists the max pixel clock at 290 MHz. The Benfei adapter has an extension block that lists a max TMDS clock of 225 MHz, but those don't seem to matter because I can hit higher without any modification. The hdmi.dat listed on the first CRU page overrides the TMDS to 340 MHz, but changing that, or even enabling DVI dual-link operation has no effect.


And I'm actually a bit surprised that more people didn't know exactly which adapter I was talking about. When I saw CommandoSnake's post, I thought there'd be huge interest in an $8 adapter that could hit at least 316 MHz. When I first got it in and was getting green artifacts all over the screen because I didn't know about the YCbCr thing, I didn't want to link people to something that didn't work.

I remember that TMDS clock of 225 MHz,the chipset of that adapter can be the Lontium LT8511,if this is true the HDMI receiver has an input bandwidth of 340 MHz and a DAC of 165 MHz,don't know if 8 or 10 bpp.
 
I did another round of WPB procedures tonight. Found something strange... maybe you guys can give some advice:

- My FW900 is very peculiar: whenever I complete the full WPB procedure start to finish, once D50 is done, the screen would flash to various modes and go into SRGB. And after I complete the process, the screen turns purple.

Usually I end the WPB procedure before the last step of D50, save the calibrated settings to a file, and manually edit the contrast back to 210. Seems like during the WPB procedure, WinDAS pushes the monitor to max contrast of 255. AFAIK the monitor is supposed to run at 210 contrast.

What I did this time was to finish the WPB process 100%, let the screen go purple. I waited another hour for the monitor (in its purple state) to warm up using a white screen. Then I ran it through another WPB procedure. And behold! The purple screen is gone and my screen looks better than ever, almost like brand new! I've had this FW900 from Unkle Vito since end of 2008, and I haven't seen it as punchy, colorful, sharp, and close to how it should be as today after running through this WPB procedures. Games look vivid and vibrant. My HD movie collection looks so sharp, I think this thing could now give my Plasma screen TV and other 4K screens a run for their money.

But three things troubles me:

1) As I've explained, after calibrating the monitor to its "purple" state, I began the WPB process all over again (without reverting back another save file) -- However, in the first step (eg step 47) of each color temperature (D93, D65, D50), I cannot get Y aka CD/M2 anywhere near the target value. All the other steps in each color mode can get to their target Y/CDM2 values just fine.

2) The G2 (and the step after G2, not sure what we can call it) values are much lower than before when I start calibrated the screen without the screen in its "purple state".

3) I'm using the i1Display Pro. It seems like this colormeter is a bit of a magnet. After I finished all my WPB process today, I noticed a blotchy green/blue area right under where the colormeter was. I was worried, until I started tinkering around and degaussed the monitor using the OSD. That cleared it right up. This has never happened before, why is it happening now? Has anyone ever experienced a colormeter or the i1Display Pro giving that much magnetic interference?
 
Last edited:
It's been a long time since I've run a WPB (I'm long overdue), but I don't think I've ever experienced this purple state. And I've never had any issue achieving target luminance in any of the steps - perhaps your cathodes aren't as responsive anymore, but that's just a guess.

As for the i1 display pro, yep, the movable part that swivels the diffuser contains a magnet, and it does cause interference. My solution is to swivel the arm only halfway - it creates a bit of an imbalance in the weight distribution when the instrument hangs down, but with a little bit of fiddling, it should hang there without causing any interference.

As for the contrast levels of 210, 255, etc. I haven't looked at them, but my understanding is that winDAS adjusts these values in order to meet certain targets. So with a new tube, a lower level of that value might produce a higher contrast. My advice is not to touch those values - doing so might have actually caused these other side effects you're experiencing, though hard to tell.
 
It's been a long time since I've run a WPB (I'm long overdue), but I don't think I've ever experienced this purple state. And I've never had any issue achieving target luminance in any of the steps - perhaps your cathodes aren't as responsive anymore, but that's just a guess.

As for the i1 display pro, yep, the movable part that swivels the diffuser contains a magnet, and it does cause interference. My solution is to swivel the arm only halfway - it creates a bit of an imbalance in the weight distribution when the instrument hangs down, but with a little bit of fiddling, it should hang there without causing any interference.

As for the contrast levels of 210, 255, etc. I haven't looked at them, but my understanding is that winDAS adjusts these values in order to meet certain targets. So with a new tube, a lower level of that value might produce a higher contrast. My advice is not to touch those values - doing so might have actually caused these other side effects you're experiencing, though hard to tell.
Thanks. Good tip about the i1Display Pro. That’s a bad design to have such strong magnets so close to the monitor.

I believe WinDAS at the end of the WPB procedure always returns the unit back to 210 contrast. That’s how it ought to be. One could manually change the dat file and run on a value higher than 210 (as many have done) but I think that’s just bad for the tube.

Today I discovered that my AG intacted FW900 has some prominent blemishes on the film that can’t be removed. So I’ll be working to remove the film these next couple of days, hopefully without trouble.

Does anyone know, in a dim or totally dark room with no direct light on the monitor, if the FW900 (after WPB) can achieve the same black level with AG off as it would with AG on?

I ask because I did read a couple of posts over the years that black level suffers after the AG film has been removed. And looking at another FW900 without AG that I have (but never got around to setting up), the bare glass looks much lighter than the AG screen. I worry that the screen can only be as dark as the bare glass; thus removing AG would hurt black levels even with proper WPB performed.
 
If you calibrate the G2 in a dimly lit or dark room, without AG, the black level will not suffer.

I measured the black level without AG in a dark room as 0.0006362 cd/m^2. That is very black.
 
If you calibrate the G2 in a dimly lit or dark room, without AG, the black level will not suffer.

I measured the black level without AG in a dark room as 0.0006362 cd/m^2. That is very black.
So after calibrating in a completely dark room, is there a big difference in black levels during the day? Just spent an hour removing my AG. Gotta wait until evening to calibrate.
 
So after calibrating in a completely dark room, is there a big difference in black levels during the day? Just spent an hour removing my AG. Gotta wait until evening to calibrate.
Yes, the blacks are terrible with lots of light in the room, unfortunately.
 
So after calibrating in a completely dark room, is there a big difference in black levels during the day? Just spent an hour removing my AG. Gotta wait until evening to calibrate.

I have no problem using the tube in my room during the day - that said, my window is behind the monitor, and I don't have bright lighting in my room.

For critical viewing, I use a bias light behind the tube - it adds a pleasant soft glow to the room and enhances the mood of the viewing experience, and the image looks wonderful.
 
I have no problem using the tube in my room during the day - that said, my window is behind the monitor, and I don't have bright lighting in my room.

For critical viewing, I use a bias light behind the tube - it adds a pleasant soft glow to the room and enhances the mood of the viewing experience, and the image looks wonderful.
I have a bias light as well. But my room during the day is probably brighter than yours so I’ll have less than good black levels during the day I suppose.

My previous USB TTL converter broke. The broken one only has three wires. Fortunately I have many spares, but they’re 4 wires (black red white and green) so I’m not sure how to plug this one into the port.

There are four pins in the FW900’s port. Do you guys know how it should be connected?

Edit: figured it out, gonna share this in case anyone has the same troubles — seems like the order from top to bottom is:

GND
+5V
RX
TX

The cable should be:
Black - GND
Red - +5V
White - RX
Green - TX

Plugging the cable in like that makes windows detect the converter but WinDAS would not work (yes I made sure WinDAS got the correct COM port).

I had to reverse the White/RX and Green/TX to get things going.
 
Last edited:
I also ended up cutting away some of the aluminum foil behind the bezel. Is this okay or is the foil there for some critical reason?
 
FYI in case this helps: I have both the Sunix and Delock adapters.

Physically, I’d say they’re the same. Outputs the same image.

However the Delock takes a second more to switch resolutions compared to the Sunix.

Also, as reported by others, the Sunix does occasionally display artifacts on the screen (wavy lines, blinking).

I’ve not seen any artifacts with the Delock adapter.
 
Yes, you need to reverse them TX and RX means transmit and receive, so TX towards RX and vice versa.

I did another round of WPB procedures tonight. Found something strange... maybe you guys can give some advice:

- My FW900 is very peculiar: whenever I complete the full WPB procedure start to finish, once D50 is done, the screen would flash to various modes and go into SRGB. And after I complete the process, the screen turns purple.

Usually I end the WPB procedure before the last step of D50, save the calibrated settings to a file, and manually edit the contrast back to 210. Seems like during the WPB procedure, WinDAS pushes the monitor to max contrast of 255. AFAIK the monitor is supposed to run at 210 contrast.

What I did this time was to finish the WPB process 100%, let the screen go purple. I waited another hour for the monitor (in its purple state) to warm up using a white screen. Then I ran it through another WPB procedure. And behold! The purple screen is gone and my screen looks better than ever, almost like brand new! I've had this FW900 from Unkle Vito since end of 2008, and I haven't seen it as punchy, colorful, sharp, and close to how it should be as today after running through this WPB procedures. Games look vivid and vibrant. My HD movie collection looks so sharp, I think this thing could now give my Plasma screen TV and other 4K screens a run for their money.

But three things troubles me:

1) As I've explained, after calibrating the monitor to its "purple" state, I began the WPB process all over again (without reverting back another save file) -- However, in the first step (eg step 47) of each color temperature (D93, D65, D50), I cannot get Y aka CD/M2 anywhere near the target value. All the other steps in each color mode can get to their target Y/CDM2 values just fine.

2) The G2 (and the step after G2, not sure what we can call it) values are much lower than before when I start calibrated the screen without the screen in its "purple state".

3) I'm using the i1Display Pro. It seems like this colormeter is a bit of a magnet. After I finished all my WPB process today, I noticed a blotchy green/blue area right under where the colormeter was. I was worried, until I started tinkering around and degaussed the monitor using the OSD. That cleared it right up. This has never happened before, why is it happening now? Has anyone ever experienced a colormeter or the i1Display Pro giving that much magnetic interference?

I have done WPB last night and I didnt see any purple screen nor I remember ever seeing one. But I am curious about the moment you left the purple screen warm up. Was that a particular step number or was it simply after you have "finished" the settings, saved them and closed WinDAS?

I am still hoping there is a way to fix the black crush without LUTs. Because LUTs are unreliable, noticeably wash out the colors (saturation issue) etc. Anyone? I tried calibrating with lights off and on, plus with some small experiments in varying WPB G2 slider values and its still there. So far the only way to prevent black crush is to raise OSD Brightness to the point where a full black pattern will show up as very noticeably grey. Is there perhaps some technical issue/limit I am missing here?
 
I am still hoping there is a way to fix the black crush without LUTs. Because LUTs are unreliable, noticeably wash out the colors (saturation issue) etc. Anyone? I tried calibrating with lights off and on, plus with some small experiments in varying WPB G2 slider values and its still there. So far the only way to prevent black crush is to raise OSD Brightness to the point where a full black pattern will show up as very noticeably grey. Is there perhaps some technical issue/limit I am missing here?

If you figure it out, let us know. I'd like to do some experimentation on that myself some time.

As for LUT reliability, they're perfectly reliable if you have a 10 bit DAC, and figure out a way to have the LUT respected across your desired range of apps. ArgyllCMS has an incredibly powerful tool (dispcal.exe) for adjusting LUTS (my instructions in the OP may need to be updated as the later versions of dispcal.exe use different switches).
 
My display also done the purple thing. I bought from Vito almost 10 years also and he told me to never use Image Restoration.

After I got my first TTL cable I dumped a dat file from the monitor and tried Image Restoration. Predicably it did something strange, it gave me a screen with a purple cast.

Once I started calibrating using WinDAS and a color meter (thanks to spacediver’s guide) I noticed that upon completing the WPB it also gave me the purple cast. The purple, I suppose, is not part of a normal WPB procedure. Just an odd issue with my unit.

So what I did is to end WPB after D50 mode has been calibrated and don’t let it go into the SRGB part of the WPB. I would then dump the dat file and turn the max contrast in the dat file back to 210 (FW900’s default, which expands to 255 during WPB).

But for some reason I recently wanted to try things again, but in a new way: I let the screen go purple upon a complete WPB procedure. Then let it warm up and run through another WPB cycle. For some reason this worked really well, giving me as far as I can recall almost the same color as when I received the unit fresh from Vito in 2008.

About crushed blacks: I had the same issue as well. I dont want to use LUT. Too much effort. I just try to use WinDAS to get the unit to optimal performance. Seems like after my “new” WPB technique, I no longer experience crushed blacks.



Yes, you need to reverse them TX and RX means transmit and receive, so TX towards RX and vice versa.



I have done WPB last night and I didnt see any purple screen nor I remember ever seeing one. But I am curious about the moment you left the purple screen warm up. Was that a particular step number or was it simply after you have "finished" the settings, saved them and closed WinDAS?

I am still hoping there is a way to fix the black crush without LUTs. Because LUTs are unreliable, noticeably wash out the colors (saturation issue) etc. Anyone? I tried calibrating with lights off and on, plus with some small experiments in varying WPB G2 slider values and its still there. So far the only way to prevent black crush is to raise OSD Brightness to the point where a full black pattern will show up as very noticeably grey. Is there perhaps some technical issue/limit I am missing here?
 
Last edited:
My display also done the purple thing. I bought from Vito almost 10 years also and he told me to never use Image Restoration.

About crushed blacks: I had the same issue as well. I dont want to use LUT. Too much effort. I just try to use WinDAS to get the unit to optimal performance. Seems like after my “new” WPB technique, I no longer experience crushed blacks.

Please set your OSD Brightness so that a full black pattern is really black (using same room lighting as during your WPB) and then open this image in Irfanview: https://i.imgur.com/ReWHzuZ.png (must be fullscreen otherwise the results are skewed) Do you see the first square at least a little bit? On my end, after normal WPB I see only square 11 and if I load LUT (from DTP94) I can see number 3.
 
FYI in case this helps: I have both the Sunix and Delock adapters.

Physically, I’d say they’re the same. Outputs the same image.

However the Delock takes a second more to switch resolutions compared to the Sunix.

Also, as reported by others, the Sunix does occasionally display artifacts on the screen (wavy lines, blinking).

I’ve not seen any artifacts with the Delock adapter.

Thanks for the report! Good to know. Hmm so I guess the Delock adapter is better then? A second more on switch resolutions is not that bad of a con.
 
Thanks for the report! Good to know. Hmm so I guess the Delock adapter is better then? A second more on switch resolutions is not that bad of a con.

I don't get wavy lines very often at all on the Sunix. And the blanking usually only happens after switching resolutions.

I'd say if you need 340+ pixel clocks, go for Sunix. If not, go Delock.
 
For new video cards with USB-C output there are also:
Delock 62796
Plugable USBC-VGA
Sunix C2VC7A0
Same chipset as Delock 62967 and the last two available in USA
And for EU users the Delock 87685,same Synaptics chip as Sunix DPU3000 for high pixel clock

Speaking about chipset,i talked with a guy from Lontium R&D center,they are updating one of their best chip and this is what he said:

We are actually updating LT8612SX to LT8612UX , which support HDMI 2.0 input, we will try to make sure that VGA output support upto 400MHz

So input bandwidth should be 600 MHz with 24 bit and 480 MHz with 30 bit,oh and DAC is 10 bit
Well it sound interesting,we'll see
 
We are actually updating LT8612SX to LT8612UX , which support HDMI 2.0 input, we will try to make sure that VGA output support upto 400MHz

So input bandwidth should be 600 MHz with 24 bit and 480 MHz with 30 bit,oh and DAC is 10 bit
Well it sound interesting,we'll see

Cool, yeah, HDMI 2.0 could possibly mean 400mHz+ pixel clocks, as long as the DAC can keep up.
 
Please set your OSD Brightness so that a full black pattern is really black (using same room lighting as during your WPB) and then open this image in Irfanview: https://i.imgur.com/ReWHzuZ.png (must be fullscreen otherwise the results are skewed) Do you see the first square at least a little bit? On my end, after normal WPB I see only square 11 and if I load LUT (from DTP94) I can see number 3.
Yes I can see the first square.
 
Yes I can see the first square.

Thank you! But wow, how is that possible? IIRC I had the same black crush on all my CRTs, on different computers. Can you please confirm you are not running any image adjustments by running the following command in CMD?

dispwin -c

During my experiments, it didnt look like different G2 values do anything to help with this. I have tried experimenting with the green central brightness or whatever its called and that only made the image greener overall. I have yet to try some other stuff there.

If you would be willing and have calibration device, what cdm2 (the "Y" value) do you get in prime mode (1920x1200@85Hz, almost fully stretched) on full white and full green patterns?

Could you dump your settings into a .dat file for me?

Sorry for bothering you, just really curious, if its too much to ask then I would totally understand if you dont have the time for this. Thanks again.
 
So I’ve removed the AG film. Screen looks sharper, brighter.

I tried to WPB so bring back the deeper blacks, however I’m thus far unable to achieve great results. If I follow the WinDAS instructions to a tee, it gives me a screen that looks too dark. Not sure what’s going on. It seems like for C_Max_B_Max, if I put the slider to where it gives me the WinDAS required CDM2 target, the screen would look dull and dark. But if I get close to the old target (what I used when I had the AG), it looks sharp. But the CDM2 reported is much higher gab what WinDAS asks for.
 
Last edited:
Thank you! But wow, how is that possible? IIRC I had the same black crush on all my CRTs, on different computers. Can you please confirm you are not running any image adjustments by running the following command in CMD?

dispwin -c

During my experiments, it didnt look like different G2 values do anything to help with this. I have tried experimenting with the green central brightness or whatever its called and that only made the image greener overall. I have yet to try some other stuff there.

If you would be willing and have calibration device, what cdm2 (the "Y" value) do you get in prime mode (1920x1200@85Hz, almost fully stretched) on full white and full green patterns?

Could you dump your settings into a .dat file for me?

Sorry for bothering you, just really curious, if its too much to ask then I would totally understand if you dont have the time for this. Thanks again.

Sure. Let me work on that.

Edit: dispwin -c doesn’t seem to work on my computer. Not sure why.
 
Last edited:
Back
Top