AMD demonstrates Freesync, G-sync equivalent?

How does AMD benefit financial from the technology? If they are giving the solution away, how is it not free?

If I set up a stand on my side walk with a "free lemonade" sign and gave away lemonade, how am I being untruthful just because you got in your car, drove to my street and incurred a cost from your use of fuel.

The way I see it is that if you want Gsync on your desktop you have to go out and buy a new Gsync monitor (at a premium over a normal monitor) and then it's free. This is something you will be able to do very soon.

If you want Freesync you have to go out and buy a new monitor with a variable refresh scaler and DP1.3 support (possibly at a premium over monitors without these technologies) as well as a new AMD GPU that supports DP1.3. Then freesync is free. You cant do that yet though since neither of those technologies are commercially available and there is no set timeline for them to become so, and indeed variable refresh rate scalers would likely never become a commercial product (at least not for a very long time) without something like Gsync to get the ball rolling in the first place.

On that basis, saying Freesync is somehow the more altruisic option or indeed simply more 'free' than Gsync seems a little strange.
 
The way I see it is that if you want Gsync on your desktop you have to go out and buy a new Gsync monitor (at a premium over a normal monitor) and then it's free. This is something you will be able to do very soon.

If you want Freesync you have to go out and buy a new monitor with a variable refresh scaler and DP1.3 support (possibly at a premium over monitors without these technologies) as well as a new AMD GPU that supports DP1.3. Then freesync is free. You cant do that yet though since neither of those technologies are commercially available and there is no set timeline for them to become so, and indeed variable refresh rate scalers would likely never become a commercial product (at least not for a very long time) without something like Gsync to get the ball rolling in the first place.

On that basis, saying Freesync is somehow the more altruisic option or indeed simply more 'free' than Gsync seems a little strange.

From my understanding there are already DP 1.2 monitors that support variable refresh. While DP 1.3 (not finalized yet) should have variable refresh as part of the standard, it is not required for variable refresh.

Regards,
SB
 
Nvidia did specifically say that to their knowledge (and in their own words - they would know) there are no monitors available today with variable refresh rate scalers (which is the key component for this).

As far as I understand it, even with DP1.3 you would still also need a monitor with a variable refresh scalar. I could be wrong there though, I must admit I'm not fully understanding the link between the scaler and DP1.3.
 
As far as I understand it, even with DP1.3 you would still also need a monitor with a variable refresh scalar. I could be wrong there though
The variable refresh rate is an optional implementation of the DP1.3 standard , monitor makers will need the proper incentive and motivation to include it, they will probably do it for high-end models under direct pressure from both AMD and NVIDIA.
 
The variable refresh rate is an optional implementation of the DP1.3 standard , monitor makers will need the proper incentive and motivation to include it, they will probably do it for high-end models under direct pressure from both AMD and NVIDIA.

You really think NVIDIA would pressure them to do it, instead of pressuring them to use G-Sync-modules instead?
 
You really think NVIDIA would pressure them to do it, instead of pressuring them to use G-Sync-modules instead?

Doesn't matter, in both cases a specific scaler has to be implemented, NVIDIA would probably use a proprietary one, while AMD will leave it to the manufacturers, customers and analysts will decide in the end whose implementation is more solid.

It could be a similar situation to 3D Vision and HD3D, one is still cared for and pretty much alive, the other one: not so much.

compatible also with earlier DisplayPort specifications. Nothing forbids the manufacturers to activate support on a DP1.2 monitor for instance.
I would have to see a model name to believe that, I also doubt it would work as flawlessly as intended, considering DP1.3 is not even finalized as of yet.
 
I would have to see a model name to believe that, I also doubt it would work flawlessly as intended, considering DP1.3 is not even finalized as of yet.
I would almost guess, that the part with the variable refresh gets more or less taken from the eDP 1.0 specs from 2008. It just defines the behaviour for a data stream (which is basically valid also under previous DP specs, just the reaction of the monitor to it is not defined so far) in that respect, which is not directly tied to any other feature or property of DP1.3. There really shouldn't be much to it to enable support also on DP1.2 interfaces (or even DP 1.0 for that matter).
I would also expect, that some scalers out there would support it already, even if it is just by accident. As long as the scaler doesn't panic if the next frame isn't coming after the exact same time as the former because vblank is asserted longer or shorter as with the previous frame (and there is no reason to panic, as long as the next refresh is coming after 33ms or 42ms, the pixel matrices of TFT panels easily hold the values that long, only after that it may get difficult for the monitor to wait longer without new values), I see no reason why it shouldn't work (at least with continuous backlight monitors, refresh locked PWM regulated ones need a bit more work). And a lot of monitors readily accept signals with very long vblank intervals (basically lowering the refresh rate at a constant pixelclock, irrespective of the interface [i.e. also with DVI or HDMI]). So the ingredients are basically there.
 
Last edited by a moderator:
Even if all the ingredients are there, I reckon they still need a good protocol (a master plan if you wish) to synchronize everything together, work at the desired precision, avoid all the possible bugs and iron out all the quirks .. activating a premature device(if it truly exists) for a giving task is usually worse than waiting for the deliberately designed one.
 
At what point have AMD said they DO exist? In fact, given that Nvidia have specifically said they don't, where is AMD refutation?

We could just go more specific about certain individuals knowing.

http://beyond3d.com/showpost.php?p=1819648&postcount=66

It's quite likely that any monitor manufacturer which has the capability but does not officially support it would not want the capability advertised.

And it isn't as if it would be difficult, in theory. If the capability already exists for laptop/notebooks, I don't see why the scaler used in those couldn't be used in a desktop monitor.

Going further there is no reason a monitor couldn't support this without a scaler in the first place. One of my HP LCD monitors doesn't have a scaler. It has 2 fixed resolutions that can be used and will display a blank screen if you attempt any other resolution. Obviously not ideal for general gaming as games that force full screen and have no option to operate at native resolution would be problematic. And this would be another situation where a monitor manufacturer wouldn't want to have their monitor advertised as supporting variable vblank.

Also the same situation where a monitor with a scaler could support variable vblank. But not in the scaler. Hence, it would only be able to do it at certain fixed resolutions. Likely just the max resolution. Again, meaning a monitor manufacturer would not want that capability divulged to the general public.

Regards,
SB
 
A competently designed scaler in a monitor takes a couple lines and outputs in real time ... it shouldn't care how long the Vblank was. Of course there is a lot of incompetence in the world.
 
The way I see it is that if you want Gsync on your desktop you have to go out and buy a new Gsync monitor (at a premium over a normal monitor) and then it's free. This is something you will be able to do very soon.

If you want Freesync you have to go out and buy a new monitor with a variable refresh scaler and DP1.3 support (possibly at a premium over monitors without these technologies) as well as a new AMD GPU that supports DP1.3. Then freesync is free. You cant do that yet though since neither of those technologies are commercially available and there is no set timeline for them to become so, and indeed variable refresh rate scalers would likely never become a commercial product (at least not for a very long time) without something like Gsync to get the ball rolling in the first place.

On that basis, saying Freesync is somehow the more altruisic option or indeed simply more 'free' than Gsync seems a little strange.

The "Free" in "FreeSync" doesnt have to be in direct relation to consumers. AMD is not providing the solution directly to gamers. Its a hardware based solution so there will be a cost to consumers paid to the manufacturers in providing the solution. That a given. But manufacturers of displays and gpu are "free" to implement the technology without being required to compensate AMD.

Its pretty obvious the "Free" in "FreeSync" has nothing to with gamers acquiring the solution at no cost but giving the market a solution that is free of proprietary control. How ubiquitous freesync becomes will be dictated by the market not by the solution provider.
 
Back
Top