Jump to content

MechWarrior 5 article from today

Recommended Posts

The MechWarrior 5 performance article makes no sense to me.  They are testing things like DLSS 2 but with an output resolution of 1080p.

The testing methodology makes knowing the real improvement impossible because DLSS 2 was really designed for 4k output and clawing back all the performance lost in running at high resolution.

I think we need a retest from someone who has at least a 2080 or 2080ti and a 4k display, otherwise most of the article is invalid and in terms of the games visual quality and performance we are missing a significant piece of the puzzle by having an output resolution below even the average which has been 1440p for at least a year based on the steam hardware survey.


*Edit* I can't reply to defend my position because the site is telling me 1 post per day is my maximum even though ive never posted before so the thread might as well be deleted since it's unfair when you can't respond to counterpoint for 24 hours.

*Edit 2*

I misspoke...wrote when I said most of the article is invalid but I can only post once per 24 hours and I can only only pm once per 24 hours so... this is going to very difficult to explain my position and why I disagree with your conclusions, especially testing at 1080p.

Edited by d0x360

Share this post

Link to post
Share on other sites

While I didnt write the article, I'm not sure where you got the idea DLSS is designed only 4K or a RTX 2080 / Ti card. 

More resolutions could be helpful, but I know for a fact it takes hours per resolution and video card. 

I did a quick look and the RTX ON with DLSS makes sense to me.

I would be interested to know other people input on the subject. Always looking to improve articles :)

Edit: Can any mod lift the 1 post restriction for the op. I haven't figured out how to.

Share this post

Link to post
Share on other sites

I did write the article and while I definitely understand a desire for more output resolutions to be tested, I lack a display with a higher resolution than 1920x1080. In theory I could use something like NVIDIA's Dynamic Super Resolution or AMD's Virtual Super Resolution, but that would actually invalidate image quality comparisons, and depending on the kind of additional processing the GPU has to do for the downscaling, the performance data might not be valid either. But also, the majority resolution according to the Steam hardware survey is 1920x1080 at 63.51%, so I'm not sure where you are seeing 2560x1440 being "the average for at least a year."

Concerning the visual quality of DLSS 2.0, perhaps you should re-read the Acceptable Qualities section where I discuss it and how significantly superior it is to the DLSS 1.0 implements I have seen before. However, I also spotted differences that made little sense, such as a loss of shadows in some areas and some missing visual effects with the laser weapons and PPC particle cannon. These differences are not present when just reducing the resolution scale, which suggests to me it is something about DLSS 2.0 and thus reasonable to expect at higher resolutions as well. Personally, if I were trying to run at a higher resolution I think I would turn to a lower resolution scale and FidelityFX sharpening, which will not suffer from the loss of shadows and other effects. Actually, that is what I did with my RX 580 in my desktop to secure better performance when playing normally, and as I stated in the article I was impressed by how well the game handles a reduced resolution scale.

To be clear, I don't mind most of your criticism. If my setup allowed for higher resolutions, I would test at them and perhaps then the DLSS results would prove more interesting to you. The comment of yours that does bother me though is claiming "most of the article is invalid" because of the lack of DLSS at higher resolutions. That is a single feature supported by only two of the eight GPUs I have. Even ray tracing is supported by more GPUs (and the GTX 1080 did a decent job with it enabled too). The purpose of these performance analysis is to see how the game runs on multiple GPUs and under different configurations when appropriate. It is not to analyze a single feature, and any article that would focus on one feature would by titled as such. At some point I would probably enjoy doing such an article, but there need to be more DLSS 2.0 supporting games, and more in my library first. I might also need a higher resolution display at that point too, at least to satisfy readers such as yourself.

As ir_cow said, hopefully the restriction can be lifted so you can reply sooner than 24 hours.

Share this post

Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

  • Create New...