Kaido
Elite Member & Kitchen Overlord
As long as the apps you use are compatible, they're not too bad!
-
We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.
The Intel Atom Thread
Page 217 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
As long as the apps you use are compatible, they're not too bad!
Kaido
Elite Member & Kitchen Overlord
Curious about ARM's future. My wife swapped out for an ARM Macbook & it's legit amazing, I think there's a lot of future potential there!
But the M1 can use nearly 40 watts under load so not quite apple to apple comparison with these processors.
Not sure about the M2.
Kaido
Elite Member & Kitchen Overlord
Oh no, not comparing it to Atom's, just saying in general, with Apple switch to ARM chips, it was actually a really good move, the performance is great!
I have an M1 in my iPad Pro & it screams! I do CAD, video editing, photo editing, 2D raster graphics, digital sculpting, etc. on it! It's crazy how far tablets have come!!
IntelUser2000
Elite Member
Per clock performance-
In order /Atom Pine Trail: 1x
Out of order/Atom Bay Trail/Braswell/Z7xxx: 1.5x
Atom Cherry Trail/Z8xxx: ~Roughly same as last gen
3-issue, out of order FP/Apollo Lake/N4200: 1.3x
Wider back end/Gemini Lake/N5000: 1.3x
Dual 3-issue/Jasper Lake/N6005: 1.3x
Enhanced Dual 3-issue, doubled FP/Gracemont: 1.3x
This doesn't take into account clock and platform related advancements. Gracemont is roughly on par with Skylake in Integer. It's not a weak core at all.
Jasper Lake gets 720-730 in GB ST in Windows and 780-790 with Linux, so with clock increase we should see something close to 1200 points. The Linux score is the one you should compare with Android devices since Android uses a Linux base.
Last edited:
Link?
Rocking a Lenovo dual core N4020 10" surface clone for RDP on the run - it can't really do anything else - I would kill to have the 4cores 6w in the same form factor.
IntelUser2000
Elite Member
Modern OSes, websites and browsers are so bloated that it takes a computer equivalent to a top tier enthusiast desktop from 10 years ago to run it properly. And the next 10 years the same thing happens.
Also I assume there's a lot of being spoiled for the performance. Because my main driver is a Kabylake-Y Core i5 4.5-7W TDP(it adjusts automatically) and it's fine.
I had the original Tablet Atom with the in-order architecture. Battery life was considered great for an x86 Tablet(7 hours) but slow. The out of order successor made everything acceptable. That's way, way slower than the N4020.
It's a combination of bad settings(like slow eMMC drive), software/OS bloat, and user expectation. Cores don't make any difference in browsing and so-called light duty tasks(since modern ones are not light at all) once you have two of them.
Roland00Address
Platinum Member
$350 today, $119 on the 27th of Nov as a Black Friday week sale (so that will be Sunday) https://www.walmart.com/ip/Gateway-...ows-10-S-Microsoft-365-Personal-1-Y/579024340
$129 https://www.walmart.com/ip/Gateway-...nt-4GB-64GB-2MP-Camera-Windows-10-S/219379468
Note these are different types of designs, one is trying to be light+big screen the sleek professional ultrabook type laptop, the other is trying to be a rugged convertible (with touchscreen) that you would want to get for a kid or for a business in the field.
Also the CPUs are not the same, one is a 4+4 of the older design, and one is 2+4 of the newer design. I am no expert of which is better, thus me asking how good these devices run as a question earlier.
I would go to SD 850 instead of 7C. is better to have 4 Big cores than 2 Big ones.
beginner99
Diamond Member
And the reason for that are abstractions. And these abstractions make sense. On one hand it is security that is greatly increased by this software abstractions. You wouldn't want to run a 80s OS on the internet. Another set of abstractions are the applications or better said programming languages themselves. Yes they are less efficient at runtime but need a fraction of the cost to develop. Doing it in C++ is more costly, even more so in C and in assembly just forget it.
I'm sure there are more abstractions and all these combined simply have a gigantic impact in performance.
Yup. What I want to see is how the GPU are performing.
Also, those i3 octa small cores, I can see it having a dGPU.
The issue this far is that nobody has been able to deliver a competitive ARM chip in the desktop or mobile space except Apple.
Microsoft should also port Windows to RISC-V because I definitely see a scenario where RISC-V overtakes ARM in the future. There are RISC-V chips launching that compete with 2016 ARM designs. That may not sound impressive, but RISC-V really didn’t start to get off the ground until a couple years ago.
That is why I think Intel needs to try harder in the embedded space. It would be neat if they could release a first party SBC with 1-2 clusters of Gracemont cores. The Pi foundation sells their capacity (something like 40,000 units) every month, so it is clear there is strong demand for such a device.
If RISC-V keeps getting faster and faster in that space, it is very likely things will trickle upstream into the desktop and server spaces.
Last edited:
Roland00Address
Platinum Member
Wouldn’t RISC V on Windows need a whole new driver thing again? Furthermore Windows wants to certify SOCs to control the platform but also maintain a basic level of performance.
Roland00Address
Platinum Member
so if that is accurate we are talking between
iPhone 7 (a10 on TSMC 16nm released 2016 with a 2+2 design)
iPhone 8 (a11 on TSMC 10nm released 2017 with a 2+4 design)
iPhone 7, Single 718, Multi 1271
iPhone 8, Single 906, Multi 1920
iPhone 14 Pro, Single 1874, Multi 5374
Hum so phone: Single 1874, Multi 5374 or pc with Single 723, multi 1493 ... which would i prefer ?
No way, this must be fake, N6000 Quad jasperlake is more powerfull than this.
Earlier leaks show over 4000 multicore score.
Yes and no. There is an N305 result on Geekbench but of course you figure that's at higher power draw.
Default string Default string vs Default string Default string - Geekbench
browser.geekbench.com
Excuse my ignorance, but the Qualcomm 8CX Gen 3 is extremely competitive against any 9W and below SoC from Intel or AMD - including those Jasper Lake and Lakemont based SoCs discussed right here.
Last edited:
I am not sure, if the Snapdragon 850 is fully compatible with Hyper-V (e.g. for Linux/WSL). On the other hand, with 4GByte memory, trying WSL or WSA (Android) is most likely pointless.
It is crazy, that you are getting LTE with these models.
Last edited:
LightningZ71
Platinum Member
I wouldn't bother with anything earlier than Gemini Lake, and gemini lake needs to be a quad with an SSD. Later chips support later Emmc standards that approach SATA speeds so, if properly specced, they can be good even with emmc. Tremont cores are reasonably useful. I think that AlderLake N though will be where everything about those chips is OK enough for most usages. The iGPU becomes performant enough to handle light gaming, there are enough threads to go around, and a single thread has enough power to keep the GDI responsive in windows.
Roland00Address
Platinum Member
Not just LTE, it is gigabit LTE if the carrier does the right thing with their 4G LTE bands allowing 10 mhz of spectrum for one person, and there is good enough cell phone towers in your area and they are not oversaturated with users.
In other words I doubt it is going to happen in real world scenarios. Not a 5G modem, but one of the better LTE modems that is now several years old.
Seeing the Ryzen 3 5300U, the Atom does not fare well.
browser.geekbench.com
The issue? AMD is going with Mendocino which is inferior than even said Ryzen (which is a direct shot on the foot)
LENOVO 82KC - Geekbench
Benchmark results for a LENOVO 82KC with an AMD Ryzen 3 5300U processor.
browser.geekbench.com
The issue? AMD is going with Mendocino which is inferior than even said Ryzen (which is a direct shot on the foot)
IntelUser2000
Elite Member
Sorry if I don't see it as little more than an excuse.
The REAL focus on security is stop and reverse the rapidly declining moral state of our world so you reduce the amount of people(not just criminals because when everyone normalizes it then what...?) that want to be vengeful.
60% of the people polled anonymously said that if they get fired/laid off they'd reveal deepest company data and break NDAs. All your security is worth nothing. Tech is worthless. Because it's about people. It always has been but we completely lost that mindset.
Meanwhile the password system is so complicated that average joes say "screw it" and just write it down next to their computers.
And how is such abstractions save time/resources in any way? When the websites have went from being few KB in size to dozens of MB and often conveying less information than they did 20 years ago?
That's why some extreme viewpoint sees all this as a ponzi scheme. Because it sounds like you artificially increase the demand of compute by making things needlessly complex.
Oh please. N300 is scoring low primarily because it's an early result. You really don't think it's the final result do you?
The N6000 Tremont top score is 2300, and that's a 6W chip. 30% on top of that is 3K.
And you would be entirely correct. Some people are just posting with feelings not their brains.
Last edited:
LightningZ71
Platinum Member
Seeing the Ryzen 3 5300U, the Atom does not fare well.
LENOVO 82KC - Geekbench
Benchmark results for a LENOVO 82KC with an AMD Ryzen 3 5300U processor.
The issue? AMD is going with Mendocino which is inferior than even said Ryzen (which is a direct shot on the foot)
5300u is a half enabled lucien/renoir chip on N7. Mendocino is essentially the same 4 core CCX with the same cache on a better node (minus some FP). I think it will be compatable at equal power.
TRENDING THREADS
-
Discussion Zen 5 Speculation (EPYC Turin and Strix Point/Granite Ridge - Ryzen 9000)- Started by DisEnchantment
- Replies: 25K
-
Discussion Intel Meteor, Arrow, Lunar & Panther Lakes + WCL Discussion Threads
- Started by Tigerick
- Replies: 25K
-
Discussion Intel current and future Lakes & Rapids thread- Started by TheF34RChannel
- Replies: 24K
-
-
