Read more.Quote:
Geekbench result from new Lenovo laptop evidence of Intel's CPU and GPU progress.
Printable View
Read more.Quote:
Geekbench result from new Lenovo laptop evidence of Intel's CPU and GPU progress.
Different version of geekbench, higher clock speed, different version of Win 10....
Seems a tad fishy to me
True. Everything from intel is fake scores. Can't trust any of it. And don't want anything from intel before they can enter the competition again.
It's geekbench.... which is now at v5 (although I wouldn't trust that much either) rather than the 2 different v4's being used here.
Not to mention (missing in top image)...
11th Gen Intel Core i7-1165G7 @ 2.80 GHz versus Intel Core i7-1065G7 @ 1.50 GHz... I'd like to think a 2.8GHz cpu would beat a 1.5GHz cpu when based on pretty similar architecture.
That clock difference would easily compensate for the difference in opencl performance between gpu's at that sort of level, even more so if it's got the 'AI' stuff added.
Exactly, always pulling scams.Quote:
Originally Posted by SpeedyJDK
I will wait until consumers have access to them and trusted sources verify exactly what chips cab and can't do.
I NEVER pay any attention to Intel propaganda.
AMD has been a little dodgy of late occasionally, too. I think it's a matter of they don't need to if Intel is being honest. But, if Intel boost their scores with trickery, then AMD feels they have to do the same. It's like the mobile SoC benchmarking stuff. If you don't do it, everyone else beats you, so you have to do it to the same extent they do.
Myself, I don't trust any of this stuff. I may have missed it but the main thing here is TDP. If you compared an 8 core AMD chip which is restricted into a 15W envelope and compare it with an Intel chip that's 4 cores, 8 threads but in a 25W envelope, you may well see these results as the single core boost may well go higher. In my mind anyway.
I wait until Anandtech has reviewed stuff before buying. It's that simple, and they have an excellent way of killing that "OOH MUST HAVE SHINY" thing with raw data. Lots of raw data. They often take time but there's a reason for that. The secondary advantage is that you get to buy later when it's cheaper.
You are making conclusions on wrong data. Read any review of the 1065G7 and it never runs at the base frequency. The Turbo frequency differences are far less between the two. The 1065G7 can Turbo to 3.9GHz in ST and 3.5GHz in MT.Quote:
Originally Posted by LSG501
Not to mention Geekbench OpenCL results do not correlate into gaming or 3D.
I also never pay attention to 'leaked' AMD benchmarks either. its to easy for a company to 'leak' a piece of completely fake info.Quote:
Originally Posted by Yoyoyo69Exactly, always pulling scams.Quote:
Originally Posted by SpeedyJDK
I will wait until consumers have access to them and trusted sources verify exactly what chips cab and can't do.
I NEVER pay any attention to Intel propaganda.
If you're going to try and correct someone at least quote their entire post.... I never even mentioned gaming and you even agree with what I was saying about there being a difference... even if it's not at exactly base (due to I assume decent cooling) there is still a difference in the frequencies. :rolleyes: Also just because a review shows something as running as higher than base frequency doesn't mean the one a person buys or one used in another review will see the same results.Quote:
Originally Posted by DavidC1
I'll quote the bit you missed on purpose...
Quote:
That clock difference would easily compensate for the difference in opencl performance between gpu's at that sort of level, even more so if it's got the 'AI' stuff added.
This nails it, if there is going to be any comparison it has to at the very least be using the same software versions.Quote:
Originally Posted by 3dcandy
...except Intel hasn't published these scores - they've simply been spotted on Geekbench.Quote:
Originally Posted by SpeedyJDK
Obviously there's too many variables to directly compare but all the article is talking about is that these results have been found on Geekbench with the quote from a current Mercedes-Benz employee that may or may not be accurate.
But to say that everything from Intel is "fake scores" is pretty much false in this context when, so far as is apparent, Intel doesn't appear to be involved in these results being published.
in general somehow suspiciously unreliable