[TR] FreeSync monitors will sample next month, start selling next year

[Mand]

"Adaptive-Sync is a standard component of VESA’s embedded DisplayPort (eDP™) specification. eDP is a companion standard to the DisplayPort interface."

So it's a "mandatory" part of eDP which is a companion standard (not mandatory) part of DP.

It's part of the standard, but not a mandatory part.

 

[Spanners]

Variable refresh isn't a mandatory part of eDP. The standard allows for it, but you don't have to implement it to use some other parts of the eDP standard.

For example, someone building a kiosk with a touchscreen might use embedded displayport, but they aren't going to care about the power saving features as much as someone using it in a battery powered device.

"The technology has been a standard component of VESA’s embedded DisplayPort (eDP&#8482 specification since its initial rollout in 2009. As a result, Adaptive-Sync technology is already incorporated into many of the building block components for displays that rely on eDP for internal video signaling. Newly introduced to the DisplayPort 1.2a specification for external displays, this technology is now formally known as DisplayPort Adaptive-Sync"

http://www.vesa.org/news/vesa-adds-adaptive-sync-to-popular-displayport-video-standard/

This was where I got my information. If "standard component" doesn't mean every eDP capable screen is Adaptive-Sync capable then I've made some incorrect assumptions. I looked into reading the entire specification but I'm not a Vesa member nor interested in paying.

Just to be clear though I said it was a mandatory (not my choice of term I was just responding to someone else, hence the quotes) part of the standard, not a companion standard like eDP is to DP. I didn't say it was mandatory to implement it on every device. Good point though I may have assumed too much.

 

[SoulWager]

"The technology has been a standard component of VESA’s embedded DisplayPort (eDP™) specification since its initial rollout in 2009. As a result, Adaptive-Sync technology is already incorporated into many of the building block components for displays that rely on eDP for internal video signaling. Newly introduced to the DisplayPort 1.2a specification for external displays, this technology is now formally known as DisplayPort Adaptive-Sync"

http://www.vesa.org/news/vesa-adds-adaptive-sync-to-popular-displayport-video-standard/

This was where I got my information. If "standard component" doesn't mean every eDP capable screen is Adaptive-Sync capable then I've made some incorrect assumptions. I looked into reading the entire specification but I'm not a Vesa member nor interested in paying.

Just to be clear though I said it was a mandatory (not my choice of term I was just responding to someone else, hence the quotes) part of the standard, not a companion standard like eDP is to DP. I didn't say it was mandatory to implement it on every device. Good point though I may have assumed too much.

There's a copy of that spec floating around on a chinese search engine, and I believe this is the relevant bit:

The Source must be capable of correctly generating all of the MSA fields, and the Sink must be capable of correctly reading all MSA fields. However, to support particular video formats from the Source, the Sink can implement the option to ignore the set of MSA Timing Parameter listed in Table 3-7 below. The Sink must be able to indicate this capability through the means described in Table 3-8.

Table 3-7: Set of MSA Timing Parameters that May Be Ignored
HTotal[15:0] HStart[15:0] HSyncPolarity (HSP) HSyncWidth[14:0] (HSW)
VTotal[15:0] VStart[15:0] VSyncPolariy (VSP) VSyncWidth[14:0] (VSW)

 

[ocre]

How it will compares to and affects gsync remains to be see. The good thing is that since it will be part of the signal spec (a sync) and when it's supported in monitors, it could potentially be utilized by all gpus (with some effort). If it's good and cheap or free it'll drive the ridiculous gsync costs down since gsync wouldn't be worth a premium.

It's clearly beneficial to consumers now we just have to wait until the implementation details and reviews start to trickle out.

Now we are getting somewhere. Its what i have been trying to get across for several post. Its not that I think negative of freesync, I really can't make such a call.

But claims that gsync is dead, that is a tremendous leap in blind faith.

We have no indication of anyone trying to use the eDP spec to create a gsync like experience and the only company saying it is even possible is AMD. They have yet to release freesync on their laptops and this telling all by itself. To use the capabilities of the spec to create a gsync experience isnt so simple. Whether anyone attempts this other than AMD is unknown. But everyone has had access to the capabilities of the new desktop spec since 2009 as it existed in the eDP spec.

So the automatic assumption that everyone will be using the desktop spec to create a gsync alternative is strange at this point as only AMD has shown interest in it.

We can imagine a future that far fetched ....
or we can bash AMD or call them a lie....

But I would like to have a real discussion on the real progress and what it actually means at this point.

 

[atticus14]

They have yet to release freesync on their laptops and this telling all by itself.

if your basing this on the laptop they demonstrated on, I'm not sure thats a good way to look at it, as it was a 4k laptop (@computex 2014 http://www.brightsideofnews.com/2014/06/07/amd-freesync-4k-laptop-computex-2014/) - my point is, being a 4k laptop it may have been one of the few that had the modern hardware to support what was likely only adaptive sync at the time, since it was running on the r9 265x which i don't think is GCN 2.0, which I think we now have established is required for AMDs additional performance gains in freesync?

Yes it's all very confusing haha. Anyway that laptop had very specific components needed for that to work, and this article provides the best explanation that I can find

http://www.anandtech.com/show/7641/amd-demonstrates-freesync-free-gsync-alternative-at-ces-2014
(take note that this is an slightly older article with a different laptop, and kinda hacked drivers I guess, so I dont think its really reasonable for AMD to support this feature without lots of possible problems on hardware thats not 100% compliant)

Regardless I don't think *Sync can just be magically turned on to existing laptops.

 

[blackened23]

Variable refresh for power savings on a mobile device? What? That sounds pretty stilly TBH because any and every mobile device that is worth a F already has significant measures in place to reduce display power during and while not in use. They will generally do half refresh rate and automatically dim the display, most ultrabooks automatically dim the display to practically nothing when the lid is closed. Every tablet that isn't a piece of junk does this. G-sync was designed to reduce tearing and input lag, and there is not a single person that moves a mouse around a screen having issues with tearing. You're moving a mouse cursor. Who cares. Most videos are sourced at 24 fps locked so no differences to be found there either. Let's face facts, this type of thing has a killer app in gaming, and no one knows how free-sync vs g-sync fares. Besides which, eDP already does its thing for power reduction for some displays on mobile devices.

However, g-sync has ULMB lightboost built in the ASIC or module. I'm not sure AMD can emulate that since lightboost was created by NV, and they have made no mention of motion blur reduction technology that can potentially ship with FS. It isn't part of any VESA spec either to my knowledge.

 

[SoulWager]

if your basing this on the laptop they demonstrated on, I'm not sure thats a good way to look at it, as it was a 4k laptop (@computex 2014 http://www.brightsideofnews.com/2014/06/07/amd-freesync-4k-laptop-computex-2014/) - my point is, being a 4k laptop it may have been one of the few that had the modern hardware to support what was likely only adaptive sync at the time, since it was running on the r9 265x which i don't think is GCN 2.0, which I think we now have established is required for AMDs additional performance gains in freesync?

Yes it's all very confusing haha. Anyway that laptop had very specific components needed for that to work, and this article provides the best explanation that I can find

http://www.anandtech.com/show/7641/amd-demonstrates-freesync-free-gsync-alternative-at-ces-2014
(take note that this is an slightly older article with a different laptop, and kinda hacked drivers I guess, so I dont think its really reasonable for AMD to support this feature without lots of possible problems on hardware thats not 100% compliant)

Regardless I don't think *Sync can just be magically turned on to existing laptops.

It wasn't working, look at the slow motion video. Fixed refresh rate.

The 4k laptop wasn't even supposed to be a demonstration of freesync, it was just a 4k laptop.

 

[VulgarDisplay]

I don't think people understand what a proof of concept or prototype is. I wonder if there was the same outrage over the wood screw GPU nvidia showed from the same individuals that are now calling freesync fake.

 

[Mand]

I don't think people understand what a proof of concept or prototype is. I wonder if there was the same outrage over the wood screw GPU nvidia showed from the same individuals that are now calling freesync fake.

Most proof of concept demonstrations usually involve, you know, proving the concept.

 

[VulgarDisplay]

Most proof of concept demonstrations usually involve, you know, proving the concept.

Explain how they didn't.

 

[SoulWager]

Explain how they didn't.

The entire point of the technology is variable refresh rate, and all the video of the demos we have is consistent with fixed refresh v-sync. At CES they used 50hz, at computex they hid the framerate numbers.

Basically, the demos needed to show significantly changing framerates. Like a slider to move it between 40 and 60fps(which is the range they claimed they had working at computex.)

Nvidia had this functionality in their very first demo, last october. AMD still hasn't demonstrated it.

 

[Eymar]

Say your panel can manage 33ms before needing to be refreshed, a refresh takes 8ms, last frame took 30ms to render, and this frame takes 35ms to render. Frames normally take close to the same amount of time to render as the previous frame, so you can reduce the risk of frame completion in the 8ms lockout by repeating the previous before you're forced to. If you repeat the frame at say 12ms, you can move your "display exactly when finished" window from 8 through 33 to 20 through 53ms.

I don't want to think about how complicated this all gets when you use multiple GPUs.

Good stuff, thanks for the detailed response. Forgot that lcd image can't persist indefinitely which if an lcd could than makes FS implementation easier. AMD implementing a "good enough" (ie. the lag may not be as good as GSync under all conditions, but if less than 16ms probably not as noticeable to many) solution looks possible. However, looks like alot more work to make Freesync as an all around smooth experience as GSync. Until AMD has a good demo of a few scenarios, really dont see GSync going away.

 

[Mand]

Explain how they didn't.

I have. Repeatedly.

Continuing to ask me to prove it then ignoring it when I do doesn't count as an argument.

 

[VulgarDisplay]

They proved that they could change the refresh rate on demand using a eDP display in a laptop. They proved their concept.

Just because you refuse to believe it doesn't make it false. At no point during their demo did they come out and say, "Look guys, freesync is here." No, they said we have this monitor with a certain eDP controller and we were able to hack it in a way that we think will allow us to mimic Gsync.

 

[SoulWager]

They proved that they could change the refresh rate on demand using a eDP display in a laptop. They proved their concept.

Just because you refuse to believe it doesn't make it false. At no point during their demo did they come out and say, "Look guys, freesync is here." No, they said we have this monitor with a certain eDP controller and we were able to hack it in a way that we think will allow us to mimic Gsync.

They did not prove anything, you can run visually identical demos on decade old fixed refresh monitors, and all the actual quantitative evidence shows a fixed 50hz refresh rate. If they proved their demos implemented variable refresh, show me the evidence of it.

Their demos showed "smooth" and "not smooth", nothing more.

 

[AtenRa]

They did not prove anything, you can run visually identical demos on decade old fixed refresh monitors, and all the actual quantitative evidence shows a fixed 50hz refresh rate. If they proved their demos implemented variable refresh, show me the evidence of it.

Their demos showed "smooth" and "not smooth", nothing more.

The demos were to show the synchronization of FPS with the monitor refresh rate (Hz) using existing technology and yes they proved that.
Yes they didnt show variable fps synchronized to variable Monitor refresh rates but that doesnt mean they lie.
Why dodnt we all wait for the next demos or the actual products before making accusations ??

 

[BrightCandle]

pcper.com spoke about that CES demo and they confirmed it was fixed at 50hz, it was not varying. They also said the more recent demo also wasn't varying either.

We haven't seen a varying refresh demo from AMD at all yet. That is unfortunately the facts.

 

[Flapdrol1337]

yeah, but the samples next month will.

I reckon it'll mostly be 60-ish hz screens without advanced overdrive (considering huddy's omg gsync has memory, full frame latency rant). looking forward to the reviews.

 

[sontin]

The Computex demo showed adaptive sync - so the GPU was matching the Hz with the actual fps - without any interaction.
Something you can do with enough buffer and eDP's PSR.

I find it more curios that VESA doesnt care to show anything about Adaptive-Sync...

 

[Abwx]

http://www.hardware.fr/html/news/cat4/page8.html#actu_13692

Avec l'Adaptive-Sync, et un support logiciel adapté, nous pourrons par exemple retrouver ce type de comportement, en toute transparence pour l'utilisateur :

- un bureau à 60 Hz
- un bureau qui se limite à 10 Hz quand il n'y a pas d'activité
- un affichage qui s'adapte aux vidéos, par exemple 24 Hz
- un affichage totalement dynamique, par exemple entre 30 et 120 Hz, dans le cadre des jeux vidéo

With Adative-sync, and an adequate software support, we could for exemple have those kind of behaviour, in all trasparency for the user :

- A desktop screen at 60hz
- A desktop screen limited to 10hz if there s no activity
- Displaying will adapt to videos, for exemple 24hz
- A totaly dynamic displaying, for exemple between 30 and 120hz, for video games

 

[sontin]

Stolen from nVidia's G-Sync marketing. You should read their "whitepaper". A simple copy of nVidia's pictures. :lol:

You know all this talk is based on G-Sync because it works. Nobody has ever seen eDP doing the same but yet they are believing AMD claims that Freesync is better than G-Sync.

 

[VulgarDisplay]

Stolen from nVidia's G-Sync marketing. You should read their "whitepaper". A simple copy of nVidia's pictures. :lol:

You know all this talk is based on G-Sync because it works. Nobody has ever seen eDP doing the same but yet they are believing AMD claims that Freesync is better than G-Sync.

I for one have been clear that I don't think freesync is better or will be better. It just needs to be good enough and cheap enough ( standard controllers mass produced) to force nvidia to support the standard and not pigeon hole their costumers into nvidia only monitors.

 

[Mand]

The demos were to show the synchronization of FPS with the monitor refresh rate (Hz) using existing technology and yes they proved that.
Yes they didnt show variable fps synchronized to variable Monitor refresh rates but that doesnt mean they lie.
Why dodnt we all wait for the next demos or the actual products before making accusations ??

My crappy work display can go from 60Hz to 50Hz. That doesn't mean it's capable of variable refresh.

No, they didn't prove that. Yes, variable fps synchronized to variable monitor refresh is what they were supposed to show and what they claimed it showed, and that is not what it showed.

We're making accusations because their demo didn't show what they claimed it showed. Why should we give them a pass and wait for the next one before judging them for presenting a faulty and deceptive demo?

The Computex demo showed adaptive sync - so the GPU was matching the Hz with the actual fps - without any interaction.
Something you can do with enough buffer and eDP's PSR.

I find it more curios that VESA doesnt care to show anything about Adaptive-Sync...

No, they didn't show variable refresh. They showed two static refresh rates.

 

[AtenRa]

My crappy work display can go from 60Hz to 50Hz. That doesn't mean it's capable of variable refresh.

No, they didn't prove that. Yes, variable fps synchronized to variable monitor refresh is what they were supposed to show and what they claimed it showed, and that is not what it showed.

We're making accusations because their demo didn't show what they claimed it showed. Why should we give them a pass and wait for the next one before judging them for presenting a faulty and deceptive demo?



No, they didn't show variable refresh. They showed two static refresh rates.

Your crappy monitor refresh rate cannot synchronize with your GPU fps output. Their Demo DID and that is evidence Freesync is working using current technology and hardware.
Granted they only did that at a steady fps rate, but the technology is working.
They also have said that more prototypes will be available in September so lets just wait and see what those prototypes can do.

 

[96Firebird]

Granted they only did that at a steady fps rate, but the technology is working.

If that is the extent of what their technology can do, I'll gladly pay a premium for Gsync.

You'd think they would want to demo a technology that can amaze us, not make us go "meh".