No, the whole reason why Lunar Lake vs M3 comparison will be so interesting is because they are both made on the same node- N3B.
So we can see who has the better architecture.
No, the whole reason why Lunar Lake vs M3 comparison will be so interesting is because they are both made on the same node- N3B.Might as well put the M4 instead of the M3, because that’s going to be its main competitor in the Apple space.
We know who has the better architecture. It’s not even close.No, the whole reason why Lunar Lake vs M3 comparison will be so interesting is because they are both made on the same node- N3B.
So we can see who has the better architecture.
@poke01 I recall you saying you wanted to create an Intel Lunar Lake vs Apple M3 thread.
The time is ripening to do so.
I won’t talk M4 here till it comes to MacBooks. iPads with M chips are truly wasted products as most users won’t do demanding work on them.Not in the MacBook yet so absolute PITA to test.
iOS is an affront to mankind, after all.
M3 vs Lunar Lake. Short version for nowNo, the whole reason why Lunar Lake vs M3 comparison will be so interesting is because they are both made on the same node- N3B.
So we can see who has the better architecture.
Agreed, but some users do real pro work on iPad. A friend of mine, who is designer, make drawings and light renderings on an iPad Pro with the stylus, before switching to his Mac. Does he need the power of an M4? Not sure, but he could do more with it.I won’t talk M4 here till it comes to MacBooks. iPads with M chips are truly wasted products as most users won’t do demanding work on them.
No, the whole reason why Lunar Lake vs M3 comparison will be so interesting is because they are both made on the same node- N3B.
So we can see who has the better architecture.
Are you referring to something like Cinebench MT per wattage?Efficiency numbers don’t look good for LNL.
Yea, we weren't expecting huge on the MT performance side.Are you referring to something like Cinebench MT per wattage?
Because at least for battery life, it seems to do pretty well from what I see of all the reviews out so far.
It was easy, just make sure to choose refind as the boot loader.I have a Lenovo Flex 2-1 14 inch convertible with the same processor. How much of a pain was CachyOS to install for you? I've got Windows 11 on it for now, and it performs well enough, but, I do like a good adventure...
Qualcomm failed hardThey delivered on the battery life claim, and the graphics are much better as well. It's everything Qualcomm said it would do, but by Intel.
Intel misled on the performance side with Lunarlake as Huang points out, but they got others right. Such is modern marketing.Qualcomm failed hard
It throttles to get that battery life.Because at least for battery life, it seems to do pretty well from what I see of all the reviews out so far.
Are you referring to something like Cinebench MT per wattage?
Both MT and 1T are significantly behind Qualcomm.They delivered on the battery life claim, and the graphics are much better as well. It's everything Qualcomm said it would do, but by Intel.
Computerbase has the 288V at 127 pts in Cinebench 2024 in ST.It throttles to get that battery life.
Both MT and 1T are significantly behind Qualcomm.
LNL has terrible overall CPU performance.
Cinebench R24 ST (Notebookcheck):Computerbase has the 288V at 127 pts in Cinebench 2024 in ST.
Computerbase has the X1E-84-100 at 127 pts.
So significantly behind is not accurate when referencing ST. At least in Cinebench.
And it’s been known the whole time that LNL wasn’t targeting MT performance.
Indeed. But that is not the point I’m making.Cinebench R24 ST (Notebookcheck):
* M3: 12.7 points/watt, 141 score
* X Elite: 9.3 points/watt, 123 score
* Intel Ultra 7 258V: 5.36 points/watt, 120 score
* AMD HX 370: 3.74 points/watt, 116 score
* AMD 8845HS: 3.1 points/watt, 102 score
* Intel 155H: 3.1 points/watt, 102 score
Taken at these power levels, X Elite has 72.9% more perf/watt while also 2.5% faster.
Cinebench R24 MT perf/watt (Notebookcheck):
* M3: 28.3 points/watt, 598 score
* X Elite: 22.6 points/watt, 1033 score
* AMD HX 370: 19.7 points/watt, 1213 score
* Intel Ultra 7 258V: 17.7 points/watt, 602 score
* AMD 8845HS: 14.8 points/watt, 912 score
* Intel 155H: 14.5 points/watt, 752 score
Taken at these power levels, X Elite has 27.7% more perf/watt while also being a whopping 71.6% faster.
Indeed. But that is not the point I’m making.
The U7 258V is not the highest SKU; the U9 288V is, which scored 127 points.
You said overall single-thread performance; not perf/watt, or IPC. That’s all.
Everything else you said is verified by the data.
Efficiency numbers don’t look good for LNL.
Where did you get the X Elite - 9.3 pts/W. The X84 scored 127, the X80 scored 123.1T is similar in performance to X Elite. However, efficiency wise, LNL is terrible. That's my main point.
I don't know how anyone can say LNL's CPU is good. It's absolutely terrible for N3B and its raw performance is severely lacking.
I think I got it from their Youtube video originally:Where did you get the X Elite - 9.3 pts/W. The X84 scored 127, the X80 scored 123.
I only see this:
View attachment 108706
Noticed the same thing — all of the indications we have other than the Geekerwan video comparing the GB16 (and using Linux) — where Andrei said the power measurements are off due to Linux — suggest LNL is pretty crappy for the area and doesn’t even live up to that much performance for the whole lineup either.Cinebench R24 ST (Notebookcheck):
* M3: 12.7 points/watt, 141 score
* X Elite: 8.3 points/watt, 123 score
* Intel Ultra 7 258V: 5.36 points/watt, 120 score
* AMD HX 370: 3.74 points/watt, 116 score
* AMD 8845HS: 3.1 points/watt, 102 score
* Intel 155H: 3.1 points/watt, 102 score
Taken at these power levels, X Elite has 54% more perf/watt while also 2.5% faster in ST. LNL has horrendous ST efficiency while also being on a better node. When AMD had these numbers over Intel, everyone here was calling Intel as dead. Now that X Elite has these numbers of Intel and you're all saying X Elite is dead.
Cinebench R24 MT perf/watt (Notebookcheck):
* M3: 28.3 points/watt, 598 score
* X Elite: 22.6 points/watt, 1033 score
* AMD HX 370: 19.7 points/watt, 1213 score
* Intel Ultra 7 258V: 17.7 points/watt, 602 score
* AMD 8845HS: 14.8 points/watt, 912 score
* Intel 155H: 14.5 points/watt, 752 score
Taken at these power levels, X Elite has 27.7% more perf/watt while also being a whopping 71.6% faster.
LNL CPU is a disaster.
Battery life of Dell XPS 13 (PC World Youtube Review):
* X Elite: 1,168 minutes, performance of 204,333 in Procyon
* LNL: 1,253 minutes in, performance of 123,000 Procyon
Battery life test with performance figures backs Cinebench results. LNL has 7% more battery life but X Elite has 65% more performance while on battery.
That's the queer thing I noticed too. Geekerwan's findings don't seem to align with those from other reviewers. There definitely is some error in the data- atleast in the SPEC2017 power curves. Geekerwan's next review video after Lunar Lake is said to be X Elite itself, so perhaps the errors might get rectified in that video.Noticed the same thing — all of the indications we have other than the Geekerwan video
GB16
GB16
Geekbench 16? Are you from the future?GB16