Intel's Arc graphics cards have quietly become excellent
Intel's Arc graphics cards have quietly become excellent
Intel's Arc A770 and A750 were decent, if imperfect, GPUs at launch, but they've massively improved since then.
"Intel’s Arc A770 and A750 were decent at launch, but over the past few months, they’ve started to look like some of the best graphics cards you can buy if you’re on a budget. Disappointing generational improvements from AMD and Nvidia, combined with high prices, have made it hard to find a decent GPU around $200 to $300 — and Intel’s GPUs have silently filled that gap.
They don’t deliver flagship performance, and in some cases, they’re just straight-up worse than the competition at the same price. But Intel has clearly been improving the Arc A770 and A750, and although small driver improvements don’t always make a splash, they’re starting to add up."
You're viewing a single thread.
Glad to see more competition in the GPU market, but the Arc cards are still not cheap enough to really take away much market share.
Userbenchmark still has the 3060ti +16% over the a770 and on the market currently the 3060ti is cheaper than the a770.
https://gpu.userbenchmark.com/Compare/Intel-Arc-A770-vs-Nvidia-RTX-3060-Ti/m1850973vs4090
I don't know much about the performance of the a770 so I don't really know if that's right, but I wouldn't trust userbenchmark at all. they favor Nvidia massively and their ratings are super inaccurate. here's an article on some of it: https://www.gizmosphere.org/stop-using-userbenchmark/
I don't trust anyone saying userbenchmark is biased without their own set of information to back up their claim. Using reddit drama as an excuse to not use a tool is weak.
The article you posted claims this:
"However, consider this: UserBenchmark mentions the NVIDIA GeForce GTX 1660 Super in their GPU section as faster than the Radeon RX 5600 XT."
This claim in the article is factually incorrect at the current time.
https://gpu.userbenchmark.com/Compare/Nvidia-GTX-1660S-Super-vs-AMD-RX-5600-XT/4056vs4062
Sounds unfair to criticize an article from 2021 for not being up to date with the ever changing metrics of Userbenchmark.
The point stands. Userbenchmark has announced and made changes to their own metric calculations because Ryzen Cpus were getting better scores than intel. It has a very clear anti-AMD stance that is clear on the written reviews and linked videos like the one in your screenshot.
Just from this comparison, I have no clue how Userbenchmark achieved the 8% given the values they're posting below. 8% is still also reasonably below what other reviewers posted at the time. https://www.techpowerup.com/review/sapphire-radeon-rx-5600-xt-pulse/27.html https://www.techspot.com/review/1974-amd-radeon-rx-5600-xt/
This just to say that they are pretty clearly biased and do not shy away from altering their metrics to favour one product over the other.
Sounds unfair to criticize an article from 2021 for not being up to date
An article from 2021 which hasn't been updated is by definition not up to date. Neither of the articles you posted even have the 3060ti on the list. Nothing you wrote actually debunks my refutation.
You attacked the credibility of the article just for having some out of date information. Just because something they reference is out of date, does not mean the article is less relevant. It's still things that happened and they should weight on how you view Userbenchmark as a source of information.
I am replying to your userbenchmark defense. Of course the articles I posted have nothing to do with the 3060ti, they were meant to source my claim that even the 8% on your posted screenshot doesn't seem like an accurate evaluation when comparing these 2 GPUs
Just because something they reference is out of date, does not mean the article is less relevant.
Yes it does.