Intel Arc A370M "Alchemist" Put Through AoTS

[btarunr]

The Intel Arc A370M was put through the "Ashes of the Singularity" benchmark. In two separate runs, the GPU ended up with a score of 3500 and 3600 points in the "Min_1080p" configuration, with an average framerate of 67 FPS in the normal batch. The notebook test-platform this chip ran on consisted of a Core i7-12700H processor, and 32 GB of memory.

3500 points is a rather vague score for this benchmark, given that GPUs from a wide range of market segments attained the similar scores (albeit on very different CPU and memory configurations). The A370M is expected to be SKU that maxes out the smaller DG2 ASIC that physically features 128 execution units (1,024 unified shaders), and a 64-bit wide GDDR6 memory bus holding 4 GB of memory. This A370M should offer roughly twice the performance as the Iris Xe iGPU with 96 EUs found in quad-core "Tiger Lake" mobile processors.



View at TechPowerUp Main Site | Source

 

[DeathtoGnomes]

maybe its good for 720p gaming?

 

[Crackong]

Oh great, they have a dGPU which is on par with competitor's last gen iGPU (And not the fastest one)

 

[bug]

Oh great, they have a dGPU which is on par with competitor's last gen iGPU (And not the fastest one)

It's a mobile part, that benchmark doesn't tell us much (if anything). The big unknown about Arc is how far can it be pushed.
Mobile parts are not a good indicator of anything, because of configurable power limits, dependency of airflow. Not to mention this particular is an engineering sample.

 

[Verpal]

Is it Ashes of the benchlarity or Bench of the Singularity?

 

[usiname]

It's a mobile part, that benchmark doesn't tell us much (if anything). The big unknown about Arc is how far can it be pushed.
Mobile parts are not a good indicator of anything, because of configurable power limits, dependency of airflow. Not to mention this particular is an engineering sample.

Enginer sample 1 week before the release? Also this benchamrk says something - A370 does not shine and old igpu can score higher. Yes A370 could have bad day, but as far as I remember I didn't saw one good result for Intel's gpus and they always have a bad day

 

[bug]

Enginer sample 1 week before the release?

Until I see the card properly identified in GPU-Z, yes, I'm going to assume engineering sample.

Also this benchamrk says something - A370 does not shine and old igpu can score higher. Yes A370 could have bad day, but as far as I remember I didn't saw one good result for Intel's gpus and they always have a bad day

It does say that about A370M, but I was saying it doesn't say anything useful about the Arc line in general.

 

[mechtech]

Wasn’t aots a thing 10 years ago???

 

[Chrispy_]

Given that this benchmark has a shitty DDR4-encumbered Vega7 IGP matching an RTX3060, I think the only information we have gained from this is that the 'benchmark' is a steaming mound of horse shit.

 

[usiname]

Given that this benchmark has a shitty DDR4-encumbered Vega7 IGP matching an RTX3060, I think the only information we have gained from this is that the 'benchmark' is a steaming mound of horse shit.

You can shoot Usain Bolt in the foot and then you will be faster than him, but this does not change the fact that Usain Bolt in good condition will out run you. You don't know what happend to this 3060 to score so low, but you can see the good results of 3060 which actually matter and show the real performance

 

[Chrispy_]

You can shoot Usain Bolt in the foot and then you will be faster than him, but this does not change the fact that Usain Bolt in good condition will out run you. You don't know what happend to this 3060 to score so low, but you can see the good results of 3060 which actually matter and show the real performance

All that proves is that every result in this 'benchmark' is an uncontrolled, unverifiable mess where no two results can be compared.

That, IMO, makes it equally worthless, just in a different and more depressing way.

 

[defaultluser]

Given that this benchmark has a shitty DDR4-encumbered Vega7 IGP matching an RTX3060, I think the only information we have gained from this is that the 'benchmark' is a steaming mound of horse shit.

No, it just means, just like all old RTS games, they have all universally become CPU-limited.

When this thing launched, in 2016, it was pushing GPU performance as-well (it was a launch DX12 title, just for PR purposes - mostly because the actual game sucked, and would have been forgotten within 15 minutes of launch!

 

[usiname]

All that proves is that every result in this 'benchmark' is an uncontrolled, unverifiable mess where no two results can be compared.

That, IMO, makes it equally worthless, just in a different and more depressing way.

So every benchamrk in the world is uncontrolled and useless because when you pick from the bottom you get absurd results
Tyme spy extreme rtx 3090 - 2000 graphics score

I scored 2 208 in Time Spy Extreme

AMD Ryzen 9 3900X, NVIDIA GeForce RTX 3090 x 1, 65536 MB, 64-bit Windows 10}

www.3dmark.com

 

[Antykain]

Wasn’t aots a thing 10 years ago???

Little late, but.. AoTS was released March 2016. It's still considered a viable stress/benchmark test due the size of the map and number of units at any given time being displayed. From what I understand, it's not the most accurate of benchmarks, but serves as a good baseline for performance. Still think AoTS needs to be retired tho.. lol

 

[mechtech]

Little late, but.. AoTS was released March 2016. It's still considered a viable stress/benchmark test due the size of the map and number of units at any given time being displayed. From what I understand, it's not the most accurate of benchmarks, but serves as a good baseline for performance. Still think AoTS needs to be retired tho.. lol

True but wasn’t early access October 2015? And I think a friend of a friend was doing closed beta testing long before that.