ive just upgraded my GPU and something doesnt feel right

[nutski1873]

Ive swapped out a 3060 12gb to a 5060ti 16gb, and on games that ive tested so far ive not really noticed much of a difference, maybe 5-10fps increase. Surely i am doing something wrong here. Anyone got any ideas as to what that might be.

 

[Tetras]

What is the rest of your PC spec?

5-10 fps from what? 30? 100?

Have you run something like hwinfo to check temps and power usage?

 

[almostgfx]

I would check the rest yeah. Could be throttling somewhere or could be a bigger issue.

I recently found out my 3600 ram was running at the default 2133 speed. That almost doubled my frames sorting that out

 

[nutski1873]

B660 Motherboard
i7 12700f
64gb ddr4 ram-3200 1600mhz 22-22-22-52
16gb MSI GeForce RTX5060ti OC Edition Ventus2x

up from depending on game really.

FH5 runs at about 75now from 65-70 before
Gears Reloaded runs 4k/high at between 80-120fps now
Fortnite - cant seem to get it above 100fps
Warcraft - from 65-70 and its now only hitting 75-80ish
Indiana Jones - 90 from 80

tried different settings but whatever i do doesnt seem to change it much. Ive tried using the Nvidia app to optimize the games or is it better just doing it in each game.

 

[Alt0153]

old ish cpu is your ram 3200 or 1600? need to play gameswith afterburner overlay enabled and see gpu usage to see if you are cpu limited

 

[throwaway4372]

What monitor?

 

[Tetras]

FH5 runs at about 75now from 65-70 before
Gears Reloaded runs 4k/high at between 80-120fps now
Fortnite - cant seem to get it above 100fps
Warcraft - from 65-70 and its now only hitting 75-80ish
Indiana Jones - 90 from 80

tried different settings but whatever i do doesnt seem to change it much. Ive tried using the Nvidia app to optimize the games or is it better just doing it in each game.

Can you show us a screenshot from hwinfo's sensor tab and run a GPU benchmark like 3D Mark. We need to check that it is getting the right scores.

 

[nutski1873]

What monitor?

It’s a Samsung Odyssey LS43CG700NUXXU

 

[melmac]

Look at the Techpowerup and TechSpot reviews for a 5060Ti. Check out the games they have and see if you have any of those games. Play those games at the settings they use and see if your FPS are similar.

 

[throwaway4372]

It’s a Samsung Odyssey LS43CG700NUXXU

Right, so that's 4K 144Hz.
I think this is about setting expectation.
Normally when someone asks what GPU to get for 4K gaming they'd be told to get a 5090, or a 5080 if they have to.

That said, 3060 12gb to 5060ti 16gb is a big upgrade so I'd expect more of a difference. It's possible there's an issue also.

Whenever someone shows up with a 4K monitor and hardware that can't really drive 4K I advise a monitor change, either 1440p or 1080p. Because usually the person has an underpowered pc due to financial constraints and that's the cheapest way to resolve it. (the alternative being spending a few grand to get something suitable for 4K)

 

[pastymuncher]

I would say your results are about right for what you are asking from it. It's not a card for 4k gaming, more like 1080p/1440p. For a decent 4k experience you need high end gpu's which is why I will never go down that road and stick with 1440p.

 

[iamgigglz]

I hate to say it but at 4K that sounds about right. I've just upgraded from a 3080Ti to a 5080 and the pure rasterized performance uplift (at 4K) was disappointing. That said, the 50 series is incredible with DLSS and frame gen and has changed my opinion of the tech completely. The newer generations of these techs look WAY better than the results I got on the 30 series card, so now I'm using them a lot more where appropriate; frame gen for slow, pretty games and DLSS for fast games where pixel detail doesn't matter, although I usually can't tell the difference with DLSS on Balanced or better.

Unless you're willing to spend £2k on a GPU you can't expect triple digit performance at 4K native.

 

[Luke_Bowman]

Unfortunately it does seem that most of the "performance uplift" of the 50 series comes from DLSS as mentioned.

Try it out and find your personal preference with the settings, there's a few modes to choose from performance, balanced, quality etc.

 

[G J]

I picked out a random 5060ti review on techpowerup and @ 4K their testing shows the 3060 12GB being nearly half the performance.

MSI GeForce RTX 5060 Ti Gaming Trio OC 16 GB Review

The MSI GeForce RTX 5060 Ti Gaming Trio OC comes with a powerful triple-fan, dual-slot cooling solution that achieves fantastic noise levels, even without a dual BIOS. Temperatures are excellent, too, and you get a decent overclock out of the box.

www.techpowerup.com

On the next page of the review it shows testing against a few cards @ 1440p one being a 3060ti and the 5060ti being 30% faster. This gap should be even higher vs a 3060 12GB.

MSI GeForce RTX 5060 Ti Gaming Trio OC 16 GB Review

The MSI GeForce RTX 5060 Ti Gaming Trio OC comes with a powerful triple-fan, dual-slot cooling solution that achieves fantastic noise levels, even without a dual BIOS. Temperatures are excellent, too, and you get a decent overclock out of the box.

www.techpowerup.com

 

[IvanDobskey]

It's just not a 4k card unfortunately. You need top end GPU to get a good performance jump on 4k. This is why I happily stay with 1440p ultra wide and mid range GPUs.

 

[Quartz]

It's just not a 4k card unfortunately.

I beg to differ: it’s not a 4K card at max settings in modern games.

 

[Pooh]

Soo...have you tried running some benchmarks to see how your set-up compares with similar set-ups to yours

 

[nutski1873]

I will run some benchmarks tmr, as stuck in bed as I can’t walk today. It’s a mad medical condition.

I know it’s not a high end 4k card, but the 3060 was running games at 4k, I just wrongly expected the 5060ti would run the same games at 4k more than 10fps better.

I will give a muck about at 1440 and see how it runs, and will play about with other settings. I do have a 32” Samsung Odyssey G5 LC32G55TQBUXXU, which is a 1440 monitor which I’ll plug in and try it out on. I’ll post the requested screenshots.