AMD Freesync Monitors & Reviews Thread

[jj109]

This image is from the Phillips 272G5DYEB Gsync monitor. It's showing ghosting. This test was done to show the effect of "SmartResponse", their overdrive. If you look at every monitor review where they do this test (There's nothing invented special here. They do them all of the time.) you'll see ghosting on every monitor, Gsync, Freesync, or not.

SOURCE

This "test" that PCPer did, has it ever been used in any other reviews? Not that I'm aware of. We've seen PCPer do this type of thing before, which is why I'm suspicious.

Your suspicions are baseless.

PCPer's comparison was done using AMD's demo which had a white windmill on a black background. On the swift with standard overdrive, there would be overdrive artifacts on the trailing edge of the windmill. However an overdriven white-to-black transition is not easily visible! I leave the reason for that as an exercise to the reader

If you have a decent monitor with okay calibration, you should be able to see the overdrive artifact on the ROG Swift in the above image.

Had the windmill demo taken place on a blue-ish background, you would have seen the overdrive artifact show as a noisy shadow behind the windmill blade.

 

[Silverforce11]

Your suspicions are baseless.

PCPer's comparison was done using AMD's demo which had a white windmill on a black background. On the swift with standard overdrive, there would be overdrive artifacts on the trailing edge of the windmill. However an overdriven white-to-black transition is not easily visible! I leave the reason for that as an exercise to the reader

If you have a decent monitor with okay calibration, you should be able to see the overdrive artifact on the ROG Swift in the above image.

Had the windmill demo taken place on a blue-ish background, you would have seen the overdrive artifact show as a noisy shadow behind the windmill blade.

So his suspicion was spot on, had PCPER used a different test, or on a different colored background, the ghosting would have been obvious. So, best case scenario for the swift was presented as "evidence" that the swift doesn't have these issues.. which is not true. These panels suffer these issues.

 

[3DVagabond]

Your suspicions are baseless.

PCPer's comparison was done using AMD's demo which had a white windmill on a black background. On the swift with standard overdrive, there would be overdrive artifacts on the trailing edge of the windmill. However an overdriven white-to-black transition is not easily visible! I leave the reason for that as an exercise to the reader

If you have a decent monitor with okay calibration, you should be able to see the overdrive artifact on the ROG Swift in the above image.

Had the windmill demo taken place on a blue-ish background, you would have seen the overdrive artifact show as a noisy shadow behind the windmill blade.

If you look at the images to the left of the pic I posted that's without any overdrive. I was simply showing ghosting. Every review I've ever seen when tested shows ghosting. This is a properly done test for ghosting not some made up test nobody has ever used with an iPhone.

 

[Destiny]

So his suspicion was spot on, had PCPER used a different test, or on a different colored background, the ghosting would have been obvious. So, best case scenario for the swift was presented as "evidence" that the swift doesn't have these issues.. which is not true. These panels suffer these issues.

Other reviews of the Rog Swift show ghosting AND "inverse ghosting" or pixel overshoot on a different color background... the Swift photo/video was taken from an angle that the "inverse ghosting" or pixel overshoot did not show up on the Freesync Windmill Demo - since it is a TN panel and color angles are not as good as an IPS.

 

[Silverforce11]

Other reviews of the Rog Swift show ghosting AND "inverse ghosting" or pixel overshoot on a different color background... the Swift photo/video was taken from an angle that the "inverse ghosting" or pixel overshoot did not show up on the Freesync Windmill Demo - since it is a TN panel and color angles are not as good as an IPS.

So its a lesson in how not to be complete amateurs and use an iPhone 6 camera & video to demonstrate panel quality.

 

[SoulWager]

Other reviews of the Rog Swift show ghosting AND "inverse ghosting" or pixel overshoot on a different color background... the Swift photo/video was taken from an angle that the "inverse ghosting" or pixel overshoot did not show up on the Freesync Windmill Demo - since it is a TN panel and color angles are not as good as an IPS.

There is visible overshoot on the ASUS in that photo, at least for the areas that were white in the previous frame. It looks black on a dark blue background, maybe it's not visible on your monitor?

edit: the background looks dark blue on my TN monitor, but dark gray on my CRT, the overshoot is still visible on both though.

 

[wand3r3r]

That's exactly why I don't trust PCPer.

They seem to be following NV's orders again, deliberately misleading (FUD), or outright incompetent if this is true.

What a joke. No wonder all of the other freesync reviews are positive and NV viral marketing is focused on ghosting to try pretend that the Gsync $150+ premium is somehow not flushing money down the drain (and ignoring the aspects of freesync which are actually superior).

It'll be an interesting year and I'm waiting to see the 4k monitors.

 

[dacostafilipe]

No wonder all of the other freesync reviews are positive and NV viral marketing is focused on ghosting to try pretend that the Gsync $150+ premium is somehow not flushing money down the drain (and ignoring the aspects of freesync which are actually superior).

To be faire, hardware.fr did also notice ghosting on their Acer XG270HU.

They did change the overdrive modes but noticed zero changes (firmware bug?).

 

[wand3r3r]

To be faire, hardware.fr did also notice ghosting on their Acer XG270HU.

They did change the overdrive modes but noticed zero changes (firmware bug?).

Yes, it is obvious it does exist to some extent on some (most/all?) of the options so far. I'd like to see (unbiased) tests between otherwise equal monitors, and knowing the color reversal technique, with different foreground and background colors..

 

[3DVagabond]

Here's an image from TFTCentral of the swift with ULMB off. Realize that you can't use ULMB and Gsync at the same time

Without ULMB there's enough blurring that you are never going to notice ghosting anyway.

Also there is ghosting and then there is artifacts from excessive overdrive. They aren't the same thing. Don't confuse them.

 

[jj109]

So his suspicion was spot on, had PCPER used a different test, or on a different colored background, the ghosting would have been obvious. So, best case scenario for the swift was presented as "evidence" that the swift doesn't have these issues.. which is not true. These panels suffer these issues.

ROG Swift suffer from overdrive artifacts which are not easily visible on a white-to-black transition. It should be obvious from the PCPer screenshot where the Swift clearly has an overshoot and the other panels aren't being overdriven enough and they ghost as a result. Overdrive artifacts != ghosting. Please get that straight.

Even if PCPer used the UFO demo, then the two Freesync panels would've looked like overdrive off samples and the swift ROG would've looked like the OD = normal samples. What would you nitpick then?

 

[jj109]

If you look at the images to the left of the pic I posted that's without any overdrive. I was simply showing ghosting. Every review I've ever seen when tested shows ghosting. This is a properly done test for ghosting not some made up test nobody has ever used with an iPhone.

But the ROG Swift doesn't ghost with standard OD settings. It has overdrive artifacts. The PCPer screenshot has overdrive artifacts on the Swift, but it's being driven to black in the demo so it's harder to see.

Is it somehow a grand PCPer and Nvidia conspiracy that AMD blessed Freesync monitors aren't being overdriven properly, and the result is really obvious when running an AMD demo?

 

[3DVagabond]

But the ROG Swift doesn't ghost with standard OD settings. It has overdrive artifacts. The PCPer screenshot has overdrive artifacts on the Swift, but it's being driven to black in the demo so it's harder to see.

Is it somehow a grand PCPer and Nvidia conspiracy that AMD blessed Freesync monitors aren't being overdriven properly, and the result is really obvious when running an AMD demo?

SOURCE

This is the Swift. Images on the left are w/o OD. OD artifacts are what you see on the right side where the yellow appears as blue in trail. I've never seen review of an LCD monitor that didn't ghost.

 

[Flapdrol1337]

Still, OD normal reduces the ghosts from 2 to 1. If the current screens don't do overdrive and freesync at the same time it's a disadvantage.

Guru3d mentioned blurring in their youtube video, oc3d said there was ghosting. But it has to be a pcper nvidia conspiracy, even though the other sites don't even bother comparing image quality?

Anyway, at least the acer is a lot cheaper, I guess I can put up with the ghosting, with 960 pixels/s I can see 8 ufo legs on my current screen

 

[3DVagabond]

They do OD and Gsync/Freesync together. They don't do ULMB (or other blur reducing tech) and variable refresh at the same time.

 

[Silverforce11]

Still, OD normal reduces the ghosts from 2 to 1. If the current screens don't do overdrive and freesync at the same time it's a disadvantage.

Guru3d mentioned blurring in their youtube video, oc3d said there was ghosting. But it has to be a pcper nvidia conspiracy, even though the other sites don't even bother comparing image quality?

Anyway, at least the acer is a lot cheaper, I guess I can put up with the ghosting, with 960 pixels/s I can see 8 ufo legs on my current screen

The difference is those sites don't claim the Swift don't ghost while that's exactly what PCPER is presenting. Which is against all evidence.

 

[Leadbox]

The red on the swift looks darker and the BenQ and LG look brighter, did pcper even try to equalise some of these differences or were they looking for a negative? I don't think the ghosting is the deal breaker they're trying to make it out to be.https://youtu.be/xjsAknzdL_Q

 

[therealnickdanger]

I really like PCPer, but they tend to give NVIDIA the benefit of the doubt, while often being in disbelief of AMD's achievements. There's a definite slant, but I don't think it goes too far beyond that.

In 3 months, we'll have the 390X, 980Ti, and a bunch more GS/FS panels to test. GAME ON!

 

[SoulWager]

I really like PCPer, but they tend to give NVIDIA the benefit of the doubt, while often being in disbelief of AMD's achievements. There's a definite slant, but I don't think it goes too far beyond that.

In 3 months, we'll have the 390X, 980Ti, and a bunch more GS/FS panels to test. GAME ON!

Sometimes it's reality that's slanted, PCPer was just the first to point out the big issues. It has been confirmed by a polish review site.(It includes video comparing the monitors at various refresh rates) http://pclab.pl/art62755-4.html

If there is a pessimistic attitude, it might be because AMD told PCPer to expect a prototype monitor in August, then beat around the bush for six months.

 

[Eymar]

Sometimes it's reality that's slanted, PCPer was just the first to point out the big issues. It has been confirmed by a polish review site.(It includes video comparing the monitors at various refresh rates) http://pclab.pl/art62755-4.html

If there is a pessimistic attitude, it might be because AMD told PCPer to expect a prototype monitor in August, then beat around the bush for six months.

The ghosting is pretty bad in those videos with only the 144hz comparison showing freesync and g-sync showing the same minimal ghosting. The link does confirm that ghosting goes away if Freesync is turned off "In addition, ghosting on the screen disappears immediately after the FreeSync function is blocked and the monitor will be able to operate with a fixed refresh rate" (Google translated link).

Hopefully drivers can fix it as suggested in the article: "The worse news is that, as we mentioned, ghosting is currently on the screens of all available monitors with FreeSync. AMD certainly will try to solve this problem in the same drivers, but in the end it may be that the company will not have access to such things as timings in the scaler board, and then the only way to eliminate ghosting can be a software update for the monitor itself, as for the end user can be very difficult."

 

[5150Joker]

That Polish website pretty much backs up what PCPer found and what Tom Petersen said.

 

[Flapdrol1337]

75 fps footage doesn't look too bad, I'd avoid the 48-75hz LG ips though.

 

[maniacalpha1-1]

What's the updated timeframe on some of these? I google, and find comments about March 2015 release dates on some, but March is slipping by...

 

[Flapdrol1337]

The BenQ has been available since early march here in europe.

 

[krumme]

That Polish website pretty much backs up what PCPer found and what Tom Petersen said.

They consistently do.