- Mar 3, 2017
- 1,777
- 6,791
- 136
It's not. They know it.And then people wonder why more and more knowledgeable people are leaving the forum.
How is "do not treat different memory configs as an apples-to-apples comparison" a controversial statement.
This pettiness seems familiar, I wonder why you created another account...Nah, I find it funny that a forum whose members overwhelmingly decided that GB6 MT is useless suddenly splitting hairs about perceived differences between submissions because reality is turning out to be quite different from the hype they have been feeding on for months.
The people who care that much use SPEC and it isn't much different in this case (except nT). GB6 is for a quick comparison.In my opinion, any geekbench ST comparison tests should be conducted with fixed clocks with all forms of boost disabled to ensure that boosting /scheduling issues won't affect the scores.
The second one is run on an Ubuntu-derivative.Heres a very interesting comparison. Someone did an all core 5GHz run on 9700X. I found an all core 5.05GHz 7800X3D to compare with. 14.7% higher ST, 11.2% higher MT.
The different geekbench versions might affect score comparison some. Hard to say. Looks pretty good though.Heres a very interesting comparison. Someone did an all core 5GHz run on 9700X. I found an all core 5.05GHz 7800X3D to compare with. 14.7% higher ST, 11.2% higher MT.
6.1-6.3 are comparable for non-SME systems according to Geekbench documentation.The different geekbench versions might affect score comparison some. Hard to say. Looks pretty good though.
Does spec also have this huge run to run variance? Kind of doubt it, really. GB workloads are obviously too short for Zen2/3/4/5 boost to reliably kick into the highest gear, which does not really happen in 'real-life' (tm) workloadsThe people who care that much use SPEC and it isn't much different in this case (except nT).
Nothing like a refreshing, new point of view to only reinforce the bad news and rumors.Looks like Zen 5% strikes again.
Yes, there is some variability in SPEC. Obviously the main source is the compiler and the optimization options which can change the score dramatically. Other factors that create variability are OS state (often official scores are run on recently booted systems), memory allocation libraries, huge TLB tweaks in the OS, etc.Does spec also have this huge run to run variance? Kind of doubt it, really. GB workloads are obviously too short for Zen2/3/4/5 boost to reliably kick into the highest gear, which does not really happen in 'real-life' (tm) workloads
Gobelins are always quickly ignored for me.And then people wonder why more and more knowledgeable people are leaving the forum.
How is "do not treat different memory configs as an apples-to-apples comparison" a controversial statement.
edit: and by "knowledgeable" I don't mean myself -- I'm far from it.
My predictions were far more accurate than Kepler/Adroc's 40% crew.Nothing like a refreshing, new point of view to only reinforce the bad news and rumors.
So fir those in the know, AMD has the 9600x and the 9700x available to buy on the 8th. So reviews HAVE to be coming Wed. But will they also cover the 9900x and the 9950x ??
No.So fir those in the know, AMD has the 9600x and the 9700x available to buy on the 8th. So reviews HAVE to be coming Wed. But will they also cover the 9900x and the 9950x ??
Very unlikely-- if MLID is right and the highest binned skus had the most validation issues (which makes sense because they were pushed back an additional week vs the lowest binned skus), then reviews on them will drop Wednesday of next week. There are still only two non ES 9950X runs on the GB browser-- no way reviews are dropping less than 48 hours from now.So fir those in the know, AMD has the 9600x and the 9700x available to buy on the 8th. So reviews HAVE to be coming Wed. But will they also cover the 9900x and the 9950x ??
Mobile lifted in July as it was not delayed.So I've been outta the game for a while (life) and came back this morning to find more than 50 pages added to this thread. Why does LTT have a review up for Strix Point if the Embargo date hasn't lifted?
Strix point embargo is lifted and laptops are available at retail now. We're discussing embargo for Granite Ridge parts, which were delayed from July 31.So I've been outta the game for a while (life) and came back this morning to find more than 50 pages added to this thread. Why does LTT have a review up for Strix Point if the Embargo date hasn't lifted?
PMC?Looks like David updated his SPEC chart includes the Ryzen 370 now.
Translation: Updated Ryzen Al 9 HX 370 (the name is really hard to pronounce) big/small core test results. The big core can touch the level of M2, and the small core is at the level of 8cx gen 3 big core. The cache capacity of these two groups is similar, but ARM has some SLC
The content tested so far does not seem to be enough to fill an article, so I'll slowly put it together after the desktop is released... Actually, PMC is quite surprising, and it will take some time to analyze.