Yep, they don't need an e core to fight MTL and Zen 4.QC may not have real E cores this round but I bet the battery life will still be great.
I think Adroc mentioned that the Zephyrus they are using for the comparisons is running single channel (64 bit) RAM....View attachment 96340
66% faster than 8945HS in PCMARK10. If some sort of benchmark manipulation trickery isn't going on, that's quite impressive.
No, they'll need it next generation even to fight AMD/Intel.Yep, they don't need an e core to fight MTL and Zen 4.
They need it to fight Apple, indirectly, in phones.
As good as Zen 5 (Strix Point) might be, it won't beat X Elite in this aspect. Even Adroc has admitted that X Elite will have fabric/idle power advantage.Look also at how low Qualcomm can go for platform power in ST — at a reduced but still functional performance of about 1/3 of each peak ST, Qualcomm can drop down to 2W whereas AMD/Intel are at around 5. Telling about the fabrics of each and the scalability of the core.
QC may not have real E cores this round but I bet the battery life will still be great.
So some say, but I think even the normal improvements and going from 8->12 cores is enough to make it a threat to SDXE.AMD is set to make a colossal leap in their P-core power efficiency with Zen 5
Look also at how low Qualcomm can go for platform power in ST — at a reduced but still functional performance of about 1/3 of each peak ST, Qualcomm can drop down to 2W whereas AMD/Intel are at around 5. Telling about the fabrics of each and the scalability of the core.
QC may not have real E cores this round but I bet the battery life will still be great.
I believe that is the only things that matters.Doesn't matter how good (performant/efficient) Intel and AMD's next gen offerings are, WoA will continue to exist and grow.
A lot of people do the majority of their day to day job inside a web browser these days. "Better browsing speed" means they can do their job faster and more easily.
Bit better but it's gonna be a lot more expensive than the Asus laptop. People really wanna pay the premium for better browsing speed?
don't worry you'll feel worse soonI am kinda disappointed with the graphics of X Elite.
I thought JavaScript has this behaviour because that's how it works by default on x86? Is this not just replicating some old SSE truncate behaviour?And ARM added instruction(s) which match JavaScript behavior to enable faster JITs: https://developer.arm.com/documentation/dui0801/h/A64-Floating-point-Instructions/FJCVTZS
With all its complexity is there even a single x64 instruction to do that? cvttsd2siq is close but still compare the implementations of TruncateDoubleToI in the arm64 and x64 macroassemblers in V8. A feature check and a single instruction vs all that crap. They just aren't innovating at the same rate.
It's close to cvttsd2siq but x64 version ends up being more instructions. At least 4 including an often untaken branch.I thought JavaScript has this behaviour because that's how it works by default on x86? Is this not just replicating some old SSE truncate behaviour?
Gory details:It's close to cvttsd2siq but x64 version ends up being more instructions. At least 4 including an often untaken branch.
Although it was much worse on ARM before that which is probably why they added it.
Ooh. Very interesting.And ARM added instruction(s) which match JavaScript behavior to enable faster JITs: https://developer.arm.com/documentation/dui0801/h/A64-Floating-point-Instructions/FJCVTZS
With all its complexity is there even a single x64 instruction to do that? cvttsd2siq is close but still compare the implementations of TruncateDoubleToI in the arm64 and x64 macroassemblers in V8. A feature check and a single instruction vs all that crap. They just aren't innovating at the same rate.
No. Not really. I smell Google. They stand to gain the most from performant Javascript. They want you to never leave ChromeOS or Android or Gmail or Google Docs. You wanna know how often I have to kill a Win32 process due to errant behavior consuming too much CPU? Almost never. How often do I kill a tab in a browser due to errant JS script eating CPU cycles? Almost every day.Ooh. Very interesting.
This also happens with FirefoxNo. Not really. I smell Google. They stand to gain the most from performant Javascript. They want you to never leave ChromeOS or Android or Gmail or Google Docs. You wanna know how often I have to kill a Win32 process due to errant behavior consuming too much CPU? Almost never. How often do I kill a tab in a browser due to errant JS script eating CPU cycles? Almost every day.
We really need independent reviews.Yeah but that doesn't explain the low 2400 ST score. Qualcomm has shown that X Elite can do 2700 in a 23W device TDP envelope.
So this Lenovo device is either a <10W fanless device or a fanned laptop but still in testing/debugging phase.
So it was you doing that. Those arrows are not for what you think they are for FYI.Btw why doesn't this thread have the functionality of upvoting/downvoting comments?
To be fair, the one thing that could pry me away from my iPad would be a 7” tablet with the x elite running Android so I agree with him.
Olrak wants an Android tablet with X Elite