External BFI Injection -- I am working with Retrotink 4K!

High Hz on OLED produce excellent strobeless motion blur reduction with fast GtG pixel response. It is easier to tell apart 60Hz vs 120Hz vs 240Hz on OLED than LCD, and more visible to mainstream. Includes WOLED and QD-OLED displays.
Post Reply
User avatar
Chief Blur Buster
Site Admin
Posts: 12296
Joined: 05 Dec 2013, 15:44
Location: Toronto / Hamilton, Ontario, Canada
Contact:

Re: External BFI Injection -- I am working with Retrotink 4K!

Post by Chief Blur Buster » 02 Oct 2026, 01:27

BFI wrote: ↑
07 Sep 2026, 23:29
Image
The Tink 4K already supports this if you disable the low-framerate BFI limit, and adjust the BFI settings. You can have

For nativeHz:frameRate ratio of 120:30, you'd be able to do the following.

- 1 visible / 3 black
- 2 visible / 2 black
- 3 visible / 1 black

Unfortunately 30Hz BFI does flicker uncomfortably. Personally I'd rather tolerate framegen lag or tolerate 60fps than 30Hz BFI. However, I strongly advocated for the ability to do 30Hz BFI for integer-divisible refresh rates (60,90,120,150,180,210,240...) which you can all do with the Tink 4K.

However, it doesn't detect the unchanged frame of a 60Hz signal, so it's crapshoot whether you're getting 1 additional frame of latency or not. That frame-synchronization feature would be so niche as not to be worth supporting (Sadly), given the limited number of people willing to tolerate 30Hz BFI. (60Hz and 50Hz BFI is already flickery, but more tolerable).

I do not recommend 30 Hz BFI if you have not pre-tested it with other software.

You can test 30 Hz BFI using TestUFO at 120 Hz.

Flicker Warning for Photosensitive Individuals
1. Set your screen to 120Hz
2. I have a hidden test, http://testufo.com/bfi-photo ...
3. Configure to 1+3, 2+2, or 3+1 for Visible+Black. That 4-frame cadence gives you a 30Hz flicker at 120Hz.
4. Stare at it for like a few minutes.
5. See how you feel afterwards

Note: If you are stuck at 240Hz (e.g. monitor does not have a 120Hz preset in Control Panel), double up the numbers 2+6, 4+4 or 6+2 to simulate BFI adjustments at 120Hz.

---

Related note.

While DLSS/FSR/etc is generally useless (currently) for esports that already maintains high frame rates, this type of game (GTA6) generally isn't nearly as latency critical. GTA6 would be an excellent candidate for framegen.

Therefore, I recommend waiting for the PC version and using one of the higher-quality framegen algorithms that will be available by its release. The current state of art in framegen appears to be DLSS 4.5 and its 6x framegen feature, when the priority is blur busting with minimum AI slop (so avoid DLSS 5 if you're more interested in temporal fidelity over spatial fidelity)

Generally in non-latency-critical games, using the "framegen compromise" can even become the overall preferable blurbusting method (than 30Hz BFI) for CP2077/GTA6 type games that aren't as latency intensive as a competitive esports game. Especially if you're playing the single player story mode.

While the "AI slop framegen" and "neural rendering" is an ongoing debate -- A better-than-average framegen on a base 60fps frame rate, for this type of game (CP2077, GTA6, Witcher, etc) now tend to produce excellent results for a 240 Hz OLED as a more ergonomic blur busting alternative to BFI, since the artifacts:fun ratio leans heavily more fun than it used to be.

Interestingly for my most recent round of CP2077 (Liberty DLC playthrough for first time while using 5x framegen)...

...I even turned off DLSS scaling and used DLSS 100% just for framegen, when playing my last round of CP2077. The artifacts of "DLSS spatial only" versus "DLSS temporal only" actually had slightly less fakeness for about 80-90% of CP2077 motion. If you're already tolerating DLSS scaling, try turning off DLSS scaling (and turn off any enhancements like DLSS neural rendering), and using only DLSS framegen if you have a low-pixel-density 240Hz OLED. It ended up being preferable!

You often notice artifacts no matter what kind of DLSS, but I preferred to milk the Hz of my 240Hz+ OLEDs, and getting motion blur reduction via brute framerate, with RTX ON. So one tactic is no DLSS supersampling, but only DLSS framegen. Set the in-game cap to the pre-multiplier pre-DLSS framerate, with a headroom to avoid the lag-yoyo effect (of VRR framerates hitting maxHz). For example, ~57fps CP2077 cap for 240Hz OLEDs for 4x DLSS.

I would assume the experience would be (roughly) similar with GTA6, assuming the DLSS framegen model doesn't glitch with GTA6 imagery, given GTA6 is also an open-world game with a story mode, though it will also have an optional online multiplayer mode (not as latency critical as FPS shooters).
Head of Blur Busters - BlurBusters.com | TestUFO.com | Follow @BlurBusters on: BlueSky | Twitter | Facebook

Image
Forum Rules wrote:  1. Rule #1: Be Nice. This is published forum rule #1. Even To Newbies & People You Disagree With!
  2. Please report rule violations If you see a post that violates forum rules, then report the post.
  3. ALWAYS respect indie testers here. See how indies are bootstrapping Blur Busters research!

roginthemachine
Posts: 24
Joined: 21 Jul 2024, 01:28

Re: External BFI Injection -- I am working with Retrotink 4K!

Post by roginthemachine » Yesterday, 01:53

Chief Blur Buster wrote: ↑
02 Oct 2026, 01:27
BFI wrote: ↑
07 Sep 2026, 23:29
Image
The Tink 4K already supports this if you disable the low-framerate BFI limit, and adjust the BFI settings. You can have

For nativeHz:frameRate ratio of 120:30, you'd be able to do the following.

- 1 visible / 3 black
- 2 visible / 2 black
- 3 visible / 1 black

Unfortunately 30Hz BFI does flicker uncomfortably. Personally I'd rather tolerate framegen lag or tolerate 60fps than 30Hz BFI. However, I strongly advocated for the ability to do 30Hz BFI for integer-divisible refresh rates (60,90,120,150,180,210,240...) which you can all do with the Tink 4K.

However, it doesn't detect the unchanged frame of a 60Hz signal, so it's crapshoot whether you're getting 1 additional frame of latency or not. That frame-synchronization feature would be so niche as not to be worth supporting (Sadly), given the limited number of people willing to tolerate 30Hz BFI. (60Hz and 50Hz BFI is already flickery, but more tolerable).

I do not recommend 30 Hz BFI if you have not pre-tested it with other software.

You can test 30 Hz BFI using TestUFO at 120 Hz.

Flicker Warning for Photosensitive Individuals
1. Set your screen to 120Hz
2. I have a hidden test, http://testufo.com/bfi-photo ...
3. Configure to 1+3, 2+2, or 3+1 for Visible+Black. That 4-frame cadence gives you a 30Hz flicker at 120Hz.
4. Stare at it for like a few minutes.
5. See how you feel afterwards

Note: If you are stuck at 240Hz (e.g. monitor does not have a 120Hz preset in Control Panel), double up the numbers 2+6, 4+4 or 6+2 to simulate BFI adjustments at 120Hz.

---

Related note.

While DLSS/FSR/etc is generally useless (currently) for esports that already maintains high frame rates, this type of game (GTA6) generally isn't nearly as latency critical. GTA6 would be an excellent candidate for framegen.

Therefore, I recommend waiting for the PC version and using one of the higher-quality framegen algorithms that will be available by its release. The current state of art in framegen appears to be DLSS 4.5 and its 6x framegen feature, when the priority is blur busting with minimum AI slop (so avoid DLSS 5 if you're more interested in temporal fidelity over spatial fidelity)

Generally in non-latency-critical games, using the "framegen compromise" can even become the overall preferable blurbusting method (than 30Hz BFI) for CP2077/GTA6 type games that aren't as latency intensive as a competitive esports game. Especially if you're playing the single player story mode.

While the "AI slop framegen" and "neural rendering" is an ongoing debate -- A better-than-average framegen on a base 60fps frame rate, for this type of game (CP2077, GTA6, Witcher, etc) now tend to produce excellent results for a 240 Hz OLED as a more ergonomic blur busting alternative to BFI, since the artifacts:fun ratio leans heavily more fun than it used to be.

Interestingly for my most recent round of CP2077 (Liberty DLC playthrough for first time while using 5x framegen)...

...I even turned off DLSS scaling and used DLSS 100% just for framegen, when playing my last round of CP2077. The artifacts of "DLSS spatial only" versus "DLSS temporal only" actually had slightly less fakeness for about 80-90% of CP2077 motion. If you're already tolerating DLSS scaling, try turning off DLSS scaling (and turn off any enhancements like DLSS neural rendering), and using only DLSS framegen if you have a low-pixel-density 240Hz OLED. It ended up being preferable!

You often notice artifacts no matter what kind of DLSS, but I preferred to milk the Hz of my 240Hz+ OLEDs, and getting motion blur reduction via brute framerate, with RTX ON. So one tactic is no DLSS supersampling, but only DLSS framegen. Set the in-game cap to the pre-multiplier pre-DLSS framerate, with a headroom to avoid the lag-yoyo effect (of VRR framerates hitting maxHz). For example, ~57fps CP2077 cap for 240Hz OLEDs for 4x DLSS.

I would assume the experience would be (roughly) similar with GTA6, assuming the DLSS framegen model doesn't glitch with GTA6 imagery, given GTA6 is also an open-world game with a story mode, though it will also have an optional online multiplayer mode (not as latency critical as FPS shooters).
Hi chief, thanks for this helpful post. I just had a look now and even when I turn “min BFI limit” to off, I can’t get 30hz flicker working with a 60hz signal into the tink, and the tink sending 120hz output to the TV.

The strobe and blur settings can only go as high as “1” each

I think the OP was referring to a way to get 60hz signals displaying 30fps content to flicker at 30hz on the tink by instructing the tink to take two 60hz frames, and only dim the latter part of the second of these. Is there definitely a way to do this on the RT4K as is, when feeding it a 60hz signal? If so, can you advise how - as strobe and blur are stuck at 1 for me.

Or do we need to try and get the tink to “half” the PS5s signal by trying a custom 30hz modeline… if that’s something the tink can do?


Thanks so much!

User avatar
Chief Blur Buster
Site Admin
Posts: 12296
Joined: 05 Dec 2013, 15:44
Location: Toronto / Hamilton, Ontario, Canada
Contact:

Re: External BFI Injection -- I am working with Retrotink 4K!

Post by Chief Blur Buster » Yesterday, 22:33

roginthemachine wrote: ↑
Yesterday, 01:53
Hi chief, thanks for this helpful post. I just had a look now and even when I turn “min BFI limit” to off, I can’t get 30hz flicker working with a 60hz signal into the tink, and the tink sending 120hz output to the TV.

The strobe and blur settings can only go as high as “1” each

I think the OP was referring to a way to get 60hz signals displaying 30fps content to flicker at 30hz on the tink by instructing the tink to take two 60hz frames, and only dim the latter part of the second of these. Is there definitely a way to do this on the RT4K as is, when feeding it a 60hz signal? If so, can you advise how - as strobe and blur are stuck at 1 for me.

Or do we need to try and get the tink to “half” the PS5s signal by trying a custom 30hz modeline… if that’s something the tink can do?

Thanks so much!
You need to do it post-deinterlace.

You need to force the 60Hz signal to 30fps using 2:2 deinterlacing, or bob/weave algorithm. Once it's internally 30fps, you can adjust BFI, as far as I can understand it.

Step 1: Deinterlace 60i -> 30p (using 2:2, or bob/weave)
Step 2: Add BFI 30p -> 60p

You need internal retrotink fps to be half output Hz, before you can add BFI.

(Incidentially, that's how 96Hz can doublestrobe with 24p material after a 3:2 deinterlace step.)
Head of Blur Busters - BlurBusters.com | TestUFO.com | Follow @BlurBusters on: BlueSky | Twitter | Facebook

Image
Forum Rules wrote:  1. Rule #1: Be Nice. This is published forum rule #1. Even To Newbies & People You Disagree With!
  2. Please report rule violations If you see a post that violates forum rules, then report the post.
  3. ALWAYS respect indie testers here. See how indies are bootstrapping Blur Busters research!

Post Reply