9% is significant
this post was submitted on 03 Jul 2025
66 points (94.6% liked)
PC Gaming
11597 readers
621 users here now
For PC gaming news and discussion. PCGamingWiki
Rules:
- Be Respectful.
- No Spam or Porn.
- No Advertising.
- No Memes.
- No Tech Support.
- No questions about buying/building computers.
- No game suggestions, friend requests, surveys, or begging.
- No Let's Plays, streams, highlight reels/montages, random videos or shorts.
- No off-topic posts/comments, within reason.
- Use the original source, no clickbait titles, no duplicates. (Submissions should be from the original source if possible, unless from paywalled or non-english sources. If the title is clickbait or lacks context you may lightly edit the title.)
founded 2 years ago
MODERATORS
TL;DW:
[1440p Average] "So across 16 games tested, the 9070 XT saw on average a 9% improvement with the latest game and driver updates. In comparison the 5070 Ti saw just a 2½% improvement."
- Steve Walton from Hardware Unboxed.
4K results shows the 5070 Ti and the 9070 XT performance matching neck and neck.
- On review day the 5070 Ti was 2% faster, but now the 9070 XT is 3% faster than the 5070 Ti.
- With the 9070 XT being $150 CAD or ~30% cheaper (in US and AU) than the 5070 Ti, the 9070 XT wins out in being the better value buy.
That's a generational upgrade according to some companies these days.
Wonder if this is similarly true on Linux platforms.
The Linux drivers were pretty buggy at launch, so probably. Every Mesa release this year has come with a ton of fixes and improvements.
I'm seeing 9070XT at €850, while a 5070Ti is at 950. This is like, 10 - 11% cheaper?
I mean, AMD is still better value, but nowhere close to 30%
Hmm. I gotta say, I got alarm bells going off in my head as soon as I saw the charts at 2:57 and 3:02. All four tests came out exactly the same in both trials? Not a single frame of difference for 1% lows or the average? I wouldn't expect results that consistent if they were run back to back, let alone months apart on new drivers...
Oh lord, 3:27 is even more suspect. The 5070 gets the exact same result twice, but then the 9070's updates make it also exactly the same??? aint no way chief
Is that based on the assumption that the numbers come from single runs instead of averages of multiple runs?
I don't see the number of runs stated anywhere. Still would be pretty unlikely to get the exact same results, and even more still for a 9070 to perform exactly the same as a 5070
So HU are fudging the numbers to favor ....?
I'm not making any claims except that these first several results look extremely suspect