RetroTINK 5x-Pro

The place for all discussion on gaming hardware
User avatar
TooBeaucoup
Posts: 398
Joined: Fri Jun 09, 2017 1:31 am

Re: RetroTINK 5x-Pro

Post by TooBeaucoup »

As someone who has played an absolute ton of games with my OSSC over the last 3 years, I rarely notice sync drops or major resolution changes. Other than the obvious heavy hitters like Silent Hill or Chrono Cross, I can't say I've run into too many instances where games are rendered unplayable or so annoying that I couldn't enjoy them. And I spend a ton of time on the PS1 and Saturn. That said, I know there's hundreds and hundreds of games, so my experiences are certainly anecdotal, but I've just never had many gripes with the OSSC. That said, my 5X is on it's way and I'm pumped! LOL!
User avatar
Tempest_2084
Posts: 593
Joined: Tue Feb 04, 2014 3:05 pm

Re: RetroTINK 5x-Pro

Post by Tempest_2084 »

I got my shipping notice today!

Image
eightbitminiboss
Posts: 450
Joined: Mon Sep 17, 2012 9:01 pm

Re: RetroTINK 5x-Pro

Post by eightbitminiboss »

TooBeaucoup wrote:As someone who has played an absolute ton of games with my OSSC over the last 3 years, I rarely notice sync drops or major resolution changes. Other than the obvious heavy hitters like Silent Hill or Chrono Cross, I can't say I've run into too many instances where games are rendered unplayable or so annoying that I couldn't enjoy them. And I spend a ton of time on the PS1 and Saturn. That said, I know there's hundreds and hundreds of games, so my experiences are certainly anecdotal, but I've just never had many gripes with the OSSC. That said, my 5X is on it's way and I'm pumped! LOL!
I've seen that the OSSC is more agreeable and tends to lock signal faster than the Framemeister so I can imagine not wanting to upgrade/sidegrade if you can help it. Especially if you play a bunch of old computers which the OSSC handles well. But man, I gotta tell you, being able to play RE2 on N64 with the expansion pak and getting to RPD without being bitten is like a dream come true on the Retrotink 5X. That game switches like crazy on screen transitions and it was absolutely maddening on a Framemeister.
User avatar
Guspaz
Posts: 3136
Joined: Tue Oct 06, 2015 7:37 pm
Location: Montréal, Canada

Re: RetroTINK 5x-Pro

Post by Guspaz »

Josh128 wrote:
Guspaz wrote: The RT5X should be always faster than the GBS-C when in framelock mode.

The extra lag from triple buffering shouldn't be enough to be noticeable at all (it's less than one frame slower than the GBS-C in the worst case), and if you're noticing a large difference, then the problem is probably with your TV switching out of game mode or adding more latency for certain input timings.

We're talking like 18ms maximum end-to-end latency with the RT5X in triple buffer mode when Bob measured it on a BVM. That's still really fast.
On CRT vs CRT for 240p>480p w/scanlines conversion with the clockgen installed, the GBS-C only adds 10 active scanlines (0.7ms or 1/24 of a frame) of lag, not 2 ms. Im very skeptical that the 5X is any faster than that, even in the framelock mode.
I was basing my numbers on Bob's latency testing on his BVM using what I'd assume is a Leo Bodnar tester, looking at his videos of both the GBS-C and the RT5X. The result was 18ms worst-case end-to-end for the RT5X using the framebuffer, and I simply don't believe that 18ms of total end-to-end latency would feel sluggish to anybody. I'd be skeptical it'd even be noticeable to anybody.
fernan1234
Posts: 2175
Joined: Mon Aug 14, 2017 8:34 pm

Re: RetroTINK 5x-Pro

Post by fernan1234 »

Guspaz wrote:I was basing my numbers on Bob's latency testing on his BVM using what I'd assume is a Leo Bodnar tester, looking at his videos of both the GBS-C and the RT5X. The result was 18ms worst-case end-to-end for the RT5X using the framebuffer, and I simply don't believe that 18ms of total end-to-end latency would feel sluggish to anybody. I'd be skeptical it'd even be noticeable to anybody.
It 100% is to someone used to CRTs. But like we've been saying, it's kind of a moot point since no one should need to use the buffer in the vast majority of cases outside of capturing.
User avatar
Guspaz
Posts: 3136
Joined: Tue Oct 06, 2015 7:37 pm
Location: Montréal, Canada

Re: RetroTINK 5x-Pro

Post by Guspaz »

That and playing any games that do the 240p/480i resolution switch, but IIRC most of those tend to be less latency sensitive games like turn-based RPGs.
fernan1234
Posts: 2175
Joined: Mon Aug 14, 2017 8:34 pm

Re: RetroTINK 5x-Pro

Post by fernan1234 »

Guspaz wrote:and playing any games that do the 240p/480i resolution switch
Even these don't need the buffer, try it for yourself and see. And even in RPGs, heck even in text adventure games, it's nice to feel near CRT-like responsiveness.
thebigcheese
Posts: 707
Joined: Sun Aug 21, 2016 5:18 pm

Re: RetroTINK 5x-Pro

Post by thebigcheese »

fernan1234 wrote:
Guspaz wrote:I was basing my numbers on Bob's latency testing on his BVM using what I'd assume is a Leo Bodnar tester, looking at his videos of both the GBS-C and the RT5X. The result was 18ms worst-case end-to-end for the RT5X using the framebuffer, and I simply don't believe that 18ms of total end-to-end latency would feel sluggish to anybody. I'd be skeptical it'd even be noticeable to anybody.
It 100% is to someone used to CRTs. But like we've been saying, it's kind of a moot point since no one should need to use the buffer in the vast majority of cases outside of capturing.
Everyone is different, I suppose, but I can say that, as someone who is very familiar with CRTs, 2 frames or less (hell, probably even 3 frames or less) of TOTAL lag is perfectly playable for me. Honestly, most modern games have more lag than that just inherent in the way they process graphics... To that end, the RT5X frame buffer is a bit of a bummer to me, but that's really only because the monitor I'm using for retro games has 2 frames of input lag (at 60 Hz), so that puts my total lag just above 3 frames. I suspect that would start to get noticeable for me. However, if I used my TV instead, which oddly has less lag, it would probably be fine.

So yes, if you are used to CRTs with no lag and then move to the buffered RT5X (~1 frame) plugged into a TV with another 2-4 frames of lag, absolutely you would notice. But on a low-lag display... nah.

And regarding the resolution switches, that's highly dependent on the display. I don't have a single display that has, with any of the upscalers I've tried (Framemeister, OSSC, RT-SCART), switched resolutions in less than 1 second. Most have taken more like 2+ seconds. It gets pretty annoying pretty fast when you need to go to the menu frequently (or whatever the thing that causes the switch is). But as Guspaz said, most of those are slower games where I would gladly take the input lag hit instead of the switching delay.
User avatar
3nemy
Posts: 4
Joined: Tue May 11, 2021 11:24 am

Re: RetroTINK 5x-Pro

Post by 3nemy »

Just a quick simple question as my order was just delivered and I'm testing all the things! :D

Does the 5x accept/understand a ypbpr signal via the scart port?

My setup includes all consoles into an Exton switcher, out to a pvm then passthrough bnc to scart. This has worked with the ossc regardless of rgbs or ypbpr (ossc let's you select ypbpr while on av1(scart)) but I'm struggling for signal with the tink.
User avatar
Guspaz
Posts: 3136
Joined: Tue Oct 06, 2015 7:37 pm
Location: Montréal, Canada

Re: RetroTINK 5x-Pro

Post by Guspaz »

No, the SCART port is RGBS and composite video only. Does the Extron switch only have one output?

Normally if people wanted to mix and match SCART and component, my suggestion would be a setup that looked something like this:

component consoles -> gcompsw -> COMP2RGB & RGB consoles -> gscartsw -> PVM & RT5X

or:

RGB consoles -> gscartsw -> RGB2COMP & component consoles -> gcompsw -> PVM & RT5X

Either using the gcompsw/gscartsw second outputs, or looping through the PVM, though I'd prefer the dual outputs to avoid any signal loss of the looping.

You could replace one of those gscartsw for a single-output extron switch or something. If it was a crosspoint, you can actually set up multi-device paths entirely on the switch, by connecting a switch output to an RGB2COMP and connecting that to a switch input, which then lets you route any input through the component conversion (or vice versa if you also have a COMP2RGB in there). But then, if it was a crosspoint, you could also just have switch outputs connected to both of the RT5X inputs.
User avatar
vol.2
Posts: 2436
Joined: Mon Oct 31, 2016 3:13 pm
Location: bmore

Re: RetroTINK 5x-Pro

Post by vol.2 »

Looks like Mike figured out the 4:2:2 V 4:4:4 thing. It was just a bug: https://twitter.com/Avaricious_Mind/sta ... 8239967232
mikechi2
Posts: 269
Joined: Sat Jul 01, 2017 6:56 am

Re: RetroTINK 5x-Pro

Post by mikechi2 »

I’m still waiting on the ultimate stamp of approval (Fudoh), but I’m hopeful I fixed the chroma bug for all modes. Can’t thank him enough for his help.
RocketBelt
Posts: 194
Joined: Thu Sep 18, 2014 8:46 pm

Re: RetroTINK 5x-Pro

Post by RocketBelt »

I looked and looked but couldn't spot the difference. Then I checked the comments and oh yeah, there's the difference.
Wow that's subtle. Chances of me spotting that during gameplay: zero.
Can't beleive you got hate for that but that's the internet for you.
mikechi2
Posts: 269
Joined: Sat Jul 01, 2017 6:56 am

Re: RetroTINK 5x-Pro

Post by mikechi2 »

idk, the people hating don't generally even look at the pics. they just hear the words "4:2:2" and are like oMG it sux. OTOH, I'm grateful for people that do zoom in and help me troubleshoot these details, which I'm keen to get right even if the differences are minor :)
User avatar
vol.2
Posts: 2436
Joined: Mon Oct 31, 2016 3:13 pm
Location: bmore

Re: RetroTINK 5x-Pro

Post by vol.2 »

Well, yeah. There is clearly an big element of mystique to all of this. It's not just people looking for a tool to get a job done. It's fanatics who view this as their central hobby and get excited about ekeing out the last little improvement or bit of performance that they can out what they have. It's like the same mentality as car people who spend a bunch of cash pointlessly squeezing an extra few horsepower out of an engine or something.

To me, the main issue is that politeness seems to be in short supply, not that people are interested in reaching a more perfect product.
notarjy
Posts: 1
Joined: Wed May 12, 2021 7:06 pm

Re: RetroTINK 5x-Pro

Post by notarjy »

Hello everyone, new here and very happy with my Retrotink 5x that I received today! I have tested so far with my RGB modded SNES Jr and N64 using HDRV component cables, and will be testing later with Gamecube using Wii component cables from the GCHDmkII. So far I have only tested using 1080p over on an old 1080p display but this weekend I will be testing with my LG CX and Dell 1440p displays as well.

I have one question regarding SNES and H. Sampling. I noticed in the menu for SD2SNES and in certain games like Seiken Densetsu 3, generic 4:3 results in clearer text compared to SNES 256. I have a feeling this is due to a high res mode of some sort, but I wanted to check if these are the expected results? I can take some pictures of the screen if needed, I dont have a direct capture device handy.

Sorry if this has been covered before, thank you.
mikechi2
Posts: 269
Joined: Sat Jul 01, 2017 6:56 am

Re: RetroTINK 5x-Pro

Post by mikechi2 »

vol.2 wrote: To me, the main issue is that politeness seems to be in short supply, not that people are interested in reaching a more perfect product.
Bingo. That and the people (really 1 or 2) that love to spread FUD for the fun of it.

To be fair, this particular issue I saw as not insignificant since it affects the input sampling and after I saw it, I couldn't unsee it even in gameplay. Mainly it caused purple fringing on the legs in DOA Xtreme Beach Volleyball. We can't have that.

The 4:2:2 on the output side, however, is far more subtle and can only be seen under extreme magnification in optimal modes.
anexanhume
Posts: 41
Joined: Sat Mar 06, 2021 1:12 am

Re: RetroTINK 5x-Pro

Post by anexanhume »

People are also extremely entitled. Consumers can yell into the void at a company’s Twitter handle and get no response. Here, you’ve got a direct line to the decision makers and he turned around a solution on a dime. In response, all you get is “Ok, what’s next?”
User avatar
orange808
Posts: 3196
Joined: Sat Aug 20, 2016 5:43 am

Re: RetroTINK 5x-Pro

Post by orange808 »

Appreciate the update and bug fixes. Maybe Micomsoft planted some messages on Twitter to help Framemeister? (Just kidding.)
We apologise for the inconvenience
User avatar
Fudoh
Posts: 13015
Joined: Mon Mar 06, 2006 3:29 am
Location: Germany
Contact:

Re: RetroTINK 5x-Pro

Post by Fudoh »

I’m still waiting on the ultimate stamp of approval (Fudoh), but I’m hopeful I fixed the chroma bug for all modes. Can’t thank him enough for his help.
happy to give my seal of approval on that one. Chroma now looks as good as it gets on the 5X.

While subtle, please don't forget that people are nowdays buying 77" or 84" TVs. 77" is ten times the area of your average 24" desktop monitor, so each pixel there equals an area larger than a 3x3 cluster on a desktop monitor.
User avatar
Kretinou
Posts: 16
Joined: Tue Jul 28, 2009 10:30 am
Location: Switzerland

Re: RetroTINK 5x-Pro

Post by Kretinou »

Got my device today and tested quickly some arcade PCBs. One word : impressive.
Do not pay attention to the colors, I did not set the RGB of my supergun correctly.

Image
PSIKYO SH2

Image
CAVE CV1000B

Image
SNK MVS

Image
Sega System 18 :shock:

Image
Taito F3 :shock: :shock: :shock:

I will test other systems tomorrow (CPS1, CPS2, System 16, System 32, ...)
ARCADE-TEAM.COM
RocketBelt
Posts: 194
Joined: Thu Sep 18, 2014 8:46 pm

Re: RetroTINK 5x-Pro

Post by RocketBelt »

Fudoh wrote:While subtle, please don't forget that people are nowdays buying 77" or 84" TVs. 77" is ten times the area of your average 24" desktop monitor, so each pixel there equals an area larger than a 3x3 cluster on a desktop monitor.
Good point! Nice work on fixing, mikechi2 8)
User avatar
evil_ash_xero
Posts: 6181
Joined: Thu Jul 12, 2007 6:33 am
Location: Where the fish lives

Re: RetroTINK 5x-Pro

Post by evil_ash_xero »

Are scanlines available in 480p? Also, is 90 comparable to the default scanlines on a 2X Pro M?
User avatar
Linkr2
Posts: 10
Joined: Wed Nov 25, 2015 1:52 pm

Re: RetroTINK 5x-Pro

Post by Linkr2 »

Hey guys! Mine arrived today and works incredible, Saturn looks like never before. Super happy with it.

One thing I noticed though, is that activating the scanlines creates a change in how the colors are displayed? Specially if you go further on 90% or 100%.

Not something super noticeable with 240p sources, but when I tried my Gamecube with Smash Bros at 480p and I activated the scanlines (I love how 480p scanlines look), I realised that some colors like the super bright yellows are generating some kind of pink interference! If you go further in the scanlines strengh, it becomes more intense.

Here are some images with the same screen showing different scanlines levels:
Spoiler
Scanlines off:
Image

Scanlines 50%
Image

Scanlines 90%
Image

Scanlines Integer 100%:
Image
I would like to know if it’s only happening for me, if someone could test it, it would be very helpful. Thanks!
Last edited by Linkr2 on Thu May 13, 2021 12:09 am, edited 3 times in total.
User avatar
Josh128
Posts: 2123
Joined: Thu Jan 16, 2014 9:01 am

Re: RetroTINK 5x-Pro

Post by Josh128 »

^
What output res are you using here? 1080p or 1440p? Looks great other than the pink stuff, lol.

Hey Mike, are you shipping the #23000s yet? Can you update on where shipping stands on launch day orders?
eightbitminiboss
Posts: 450
Joined: Mon Sep 17, 2012 9:01 pm

Re: RetroTINK 5x-Pro

Post by eightbitminiboss »

Linkr2 wrote:Hey guys! Mine arrived today and works incredible, Saturn looks like never before. Super happy with it.

One thing I noticed though, is that activating the scanlines creates a change in how the colors are displayed? Specially if you go further on 90% or 100%.

Not something super noticeable with 240p sources, but when I tried my Gamecube with Smash Bros at 480p and I activated the scanlines (I love how 480p scanlines look), I realised that some colors like the super bright yellows are generating some kind of pink interference! If you go further in the scanlines strengh, it becomes more intense.

Here are some images with the same screen showing different scanlines levels:
Spoiler
Scanlines off:
Image

Scanlines 50%
Image

Scanlines 90%
Image

Scanlines Integer 100%:
Image
I would like to know if it’s only happening for me, if someone could test it, it would be very helpful. Thanks!
I'll check when I get home tonight, but I don't recall having this issue with Eternal Darkness.
mikechi2
Posts: 269
Joined: Sat Jul 01, 2017 6:56 am

Re: RetroTINK 5x-Pro

Post by mikechi2 »

There's probably a bug in the scanline generation that needs to be looked @. TBH, I didn't imagine anyone would use them for 480p content.

23000s are the very end, so it might be a bit longer. Not sure if it makes you feel better but you were probably the last ones to make it. Thanks for your patience!
User avatar
FinalBaton
Posts: 4461
Joined: Sun Mar 08, 2015 10:38 pm
Location: Québec City

Re: RetroTINK 5x-Pro

Post by FinalBaton »

You truly are conducting business well so far Mike, the community is lucky to have you. You're quick to fix the important issues/bugs and are open to add features that can benefit many use cases. Thank you for your work, sir.

Of course, people need to be polite with you. Let's hope the impolite cases will be even fewer and far between from now on.



btw it was interesting listening your live on RetroRGB. it became obvious to me that 4:2:2 at the input was a non-issue since it's oversampled. a perfectly fine design compromise, as you've said. even with the tiny chroma thing it looked fantastic, but looks like you've ironed it out now! wow
-FM Synth & Black Metal-
User avatar
Josh128
Posts: 2123
Joined: Thu Jan 16, 2014 9:01 am

Re: RetroTINK 5x-Pro

Post by Josh128 »

Mike-- I think those 480p scanlines look fantastic (colors not withstanding :mrgreen: ), but Im unclear as to the resolution used here. It must be 1440p Im guessing? Or 960p in a 1080p frame?

As far as the order, I lucked out indeed, my son was uploading a gigantic file from his room, unbeknownst to me, and was slowing the hell out of my internet connection speed. I had a couple of failed attempts to purchase the unit before it went through. Im really curious to try some GC out on my 1080p plasma and 240p systems on my 768p one.
User avatar
Linkr2
Posts: 10
Joined: Wed Nov 25, 2015 1:52 pm

Re: RetroTINK 5x-Pro

Post by Linkr2 »

Thanks mikechi2, it seems so.

The previous ones I uploaded were at 1080p UNDER, and this new ones are at 1440p, both looking amazing picture quality wise and scanlines wise (obviating the color issue). Here are the new ones showing again this possible bug in the scalines generation:
Spoiler
Scanlines off:
Image

Scanlines 50%
Image

Scanlines 90%
Image

Scanlines Integer 100%:
Image
I will try to provide more images from other systems later today. If the images are too big please let me know, I am uploading them this way to better picture what is happening and so the scanlines are not ruined by shrinking them. Thanks.
Post Reply