Pimax Please Make FOV Reduction OPTIONAL in Future Versions of Pitools

Pimax Please Make FOV Reduction OPTIONAL in Future Versions of Pitools
0

I made a previous thread in which I said Pimax shrank vertical FOV in firmware 255.

I was wrong. They didn’t reduce it in the firmware. They reduced it in Pitools itself.

This is easily verifiable by reverting to Pitools 1.0.1.245_R255 or below.

I first noticed this when updating from the Pitools .180 beta to .253. I saw what looked like an obvious drop in vertical FOV. But no one else seemed to be talking about it, so I wondered if perhaps I’d imagined it.

But I wasn’t the only one who saw this. User grodenglaive on the Pimax reddit forum had also noticed it too, and he suggested I try Pitools 2.45, because it still had the larger vertical FOV. So I installed .245–and immediately saw vertical FOV increase back to the level I’d seen before updating from the .180 beta! (In addition to shrinking vertical FOV, it looks like they changed the overall shape of the
view, and this makes the edges a lot more noticeable in general, IMO).

I don’t think Pimax should reduce FOV, because increased FOV is the main reason I went with Pimax in the first place, and I know I’m not the only one. At the very least they should acknowledge that they have reduced vertical FOV and make this reduction optional in future updates.

Before anyone weighs in to suggest that this decrease is due to the new higher refresh rates and can be fixed by dropping back down to 90hz, this is simply incorrect. Nor can it be fixed by disabling Hidden Area Mask or any other setting. It is simply a software reduction of vertical FOV in Pitools itself, and anyone doubting this can revert back to .245 or .180, etc–and see for yourselves.

Pimax–PLEASE MAKE THIS REDUCTION OPTIONAL in future versions of Pitools.

5 Likes

A good catch. I believe this could be related to the update applied in .197?

‘Physical resolution reduction
–The adaptive physical limitation of panel pixel area utilization when FOV is reduced. Approx 10% performance / framerate increase in VR.’

I could be totally wrong as I’m going off of memory from months ago, but I remember @risa2000 (at least I think it was risa!) bringing up that the geometric math didn’t make sense, and that extra pixels were being rendered that weren’t even displayed on the panels. Perhaps this optimization went too far? I could be totally wrong, just a thought.

4 Likes

@MyNameIs do you talk about the Parallel projection mode, or the “native” one?

One way to confirm that could be by looking at the same scene (image) by using “display VR view” in Steam (if it is possible with Pimax) and saving the screenshots for the two versions of PiTool and then compare them.

The other way is by using my tool (https://github.com/risa2000/hmdq) and comparing the new values with the database here (https://risa2000.github.io/hmdgdb/). This method however does not show easily if there is a difference in the “hidden area mask”, if Pimax changed it.

@Thall If you had my proposal for “HAM mod” on mind (Pimax 5k+ parallel projection HAM mod) then this change should only affect the PP on mode.

2 Likes

A good catch. I believe this could be related to the update applied in .197?

I think you’re right. Although the meaning of “adaptive physical limitation” isn’t exactly clear, they seem to be stating outright that they did reduce FOV to improve performance.

That’s a completely unacceptable trade off when it is forced on the end user rather than being presented as an option.

Imagine if Pimax announced they were improving performance … by reducing resolution.

“Hey Pimax owners. You know that resolution you had? We decided to reduce that a little for you. You don’t have any say in this, but you do get a higher frame rate, and we don’t have to find another way to optimize performance. It’s win win.”

People would go absolutely berserk. No one would accept this.

It’s no different with FOV. The reason I paid a lot of money for this specific HMD is because it offered a specific FOV. That’s why I bought it. I don’t want to give up any FOV for more FPS. Let Pimax owners decide if it’s better to trade FOV for performance. Don’t impose this decision upon us.

3 Likes

100% agree.

I have suggested myself an option to be able to adjust vertical FOV for performance gain but this must be a decision made by the user and not something imposed, especially if this could be some hidden move to make people think performance gains have been made through “optimizations” hoping nobody will notice the V-Fov has been silently reduced.

1 Like

Wasnt the FOV automatically reduced in 120hz mode due to bandwidth limitations?

I did notice i change when going to 120hz, i think ive gone up to 120 large which i feel is quite close to 90 normal iirc.

1 Like

I’ve also noticed this - I noticed the difference immediately. I’m hoping to return to all of the display

1 Like

Is it reduced in 90Hz mode as well?? :astonished:

Yeah, I want to wait to show my friends because the vertical FOV was really nice

1 Like

@SweViver @PimaxQuorra

Could you please have a look at this ? :slight_smile:

2 Likes

Wasnt the FOV automatically reduced in 120hz mode due to bandwidth limitations?

Yes it was, but this has nothing to do with that. This is a visible reduction in vertical FOV across the board no matter what refresh rate you set. I have an XR which only goes up to 82hz and vertical FOV has been reduced in that mode too.

Is it reduced in 90Hz mode as well??

My OLED panel only goes up to 82hz but since Pimax cut vertical FOV in that mode it seems like a sure bet that they cut it in higher modes too, and others with LCD panels and higher refresh rates are reporting the same thing.

Could you please have a look at this ?

Yes, please help us fix this.

The desire for higher FOV is the core motivation that gave rise to Pimax in the first place. We can’t go backwards from what we already had. Or at the very least it should be optional. If some owners want to trade reduced FOV for higher FPS, let them (although I don’t think they even realized that was the deal they were getting). But it shouldn’t be imposed on the rest of us. Let us have the choice.

If you Run @risa2000 hmdq program it will give you hard values of requested vertical & Horizontal rendered FoV.

1 Like

Findings with Risa2000’s tool

5K+ FW 255
Pitool 197
120Hz Large FOV

horizontal: 132.39 deg
vertical: 103.03 deg
diagonal: 135.63 deg
overlap: 86.79 deg

5K+ FW 255
Pitool 249
120Hz Normal FOV

Total FOV
horizontal: 130.87 deg
vertical: 103.03 deg
diagonal: 134.87 deg
overlap: 86.79 deg

5K+ FW 255
Pitool 255
120Hz Normal FOV

Total FOV
horizontal: 127.66 deg
vertical: 103.03 deg
diagonal: 124.96 deg
overlap: 86.79 deg

5K+ FW 255
Pitool 258
120Hz Normal FOV

Total FOV
horizontal: 127.66 deg
vertical: 103.03 deg
diagonal: 124.95 deg
overlap: 86.79 deg

Interresting indeed… :slight_smile:
Maybe we should tests earlier firmwares to confirm if the vertical FOV has been altered in that.

2 Likes

This one looks suspicious, I would not expect diagonal to be smaller than horizontal, unless there is some big hidden area mask masking the corners of the FOV.

If you would have a chance to get into this particular config again (or the one for PiTool 258), could you send me the full data dump, which you get by?

hmdq -n --out_json 5K_FW255_P258_Normal_120Hz.json
2 Likes

Sure :slight_smile:

I have the one for FW 255 and PiTool 255
http://ahlefeld.dk/5K_FW255_P255_Normal_120Hz.json

3 Likes

I rendered the visualization and the numbers seem legit (considering the hidden area mask).

3 Likes

risa2000, could you please explain how to use your tool for someone whose really dumb? I don’t, uh, mean myself, I mean my friend, who is a little confused about how to use your tool. Taking screenshots through Steam doesn’t seem to work for Pimax.

Well, the thorough “user manual” is directly on the GitHub page (https://github.com/risa2000/hmdq) just scroll down to “README”. If your friend however feels overwhelmed by this, then I guess the sufficient minimum to start (hopefully :wink:), is to acknowledge that:

  • Get the tool here (https://github.com/risa2000/hmdq/releases/latest)
  • Unzip it into a convenient folder.
  • It is a command line tool, so one needs to run it from a command line.
  • Get to the command line (Windows console) by pressing Windows key + R and then type in cmd.exe.
  • This will start a “console” window, most likely in the user’s home directory.
  • Type cd /d <convenient folder you chose above> to switch to the folder with the tool.
  • Type hmdq.exe to see the basic information about the headset geometry (SteamVR must be running and the headset connected/recognized).
  • Type hmdq.exe help for other options or read the manual :wink:.

What you will see will be the FOV reported by the headset to SteamVR and then by SteamVR to hmdq tool, so basically the same figures any other app, which uses SteamVR sees.

3 Likes