Samsung Exynos 5250 - production starting in Q2 2012

  • Thread starter Deleted member 13524
  • Start date
Turbotab said:
Regarding the CPU scores, I wonder if Samsung are using HMP, because if the scores include the 4 x A53s then I'm not overly impressed by the A57s, unless they are clocked very low, which would raise a whole load of new questions.
Click to expand...
Forget that A53/A57 nonsense. It's a plain A7/A15 chip. I still wonder how websites can still be reporting on the 64 bit shenanigans that came up late last year regarding the 5430.

And yes it's a HMP chip.
 
Nebuchadnezzar said:
Forget that A53/A57 nonsense. It's a plain A7/A15 chip. I still wonder how websites can still be reporting on the 64 bit shenanigans that came up late last year regarding the 5430.

And yes it's a HMP chip.
Click to expand...

http://www.gsmarena.com/both_samsung_galaxy_note_4_versions_visit_antutu-news-8826.php

There is a screenshot of the Antutu info page, which states A57/A53, so this is either a forgery, Antutu is fubar or it is a A57 SoC. A quick comparison of the S5 A15/A7 Antutu CPU scores indicates that you're right and this is a slightly highly clocked A15/A7, either that or A57 is rubbish.
 
Turbotab said:
http://www.gsmarena.com/both_samsung_galaxy_note_4_versions_visit_antutu-news-8826.php

There is a screenshot of the Antutu info page, which states A57/A53, so this is either a forgery, Antutu is fubar or it is a A57 SoC. A quick comparison of the S5 A15/A7 Antutu CPU scores indicates that you're right and this is a slightly highly clocked A15/A7, either that or A57 is rubbish.
Click to expand...
All Antutu does it match the SoC with a database and match a core description to that SoC ID. They don't even look at the CPUID registers of the chips like CPU-Z or others do. If that database is filled with nonsense, it will report nonsense in the app.

The source code describing the CPUs has been out for months: https://github.com/AndreiLux/UNIVERSAL5422/blob/exynos5422/arch/arm/boot/dts/exynos5430.dtsi#L57

It's really not worth arguing about it, it's wrong and one should just ignore these reports.
 
Ailuros said:
You might want to have a look at AndEBench Pro as a SoC benchmark.
Click to expand...
Just another benchmark that is incapable of correctly reading out CPUFreq policies. You would think *one* of the developers of these benchmarks over the last couple of years would be able to read Linux's documentation on the matter.
 
No idea about what its reading honestly and I only had a very quick look at it. I meant it more as an alternative SoC benchmark to Antutu.
 
Exynos 3250 (2x A7 1GHz Mali 400MP2 133MHz)
Exynos 3470 (4x A7 1.4GHz, Mali 400MP4 450MHz)
Exynos 4415 (4x A9 1.4GHz, Mali 400MP4 533MHz, integrated LTE modem)
Exynos 5250 (2x4 A15/A7, T624 600MHz)
Exynos 5412 (4x4 A15/A7, T624 (??) 600MHz)
Exynos 5422 (4x4 A15/A7, T628MP6 533MHz)
Exynos 5430 (4x4 A15/A7, T760 600MHz)
Exynos 5433 (4x4 A15/A7, T760 ? MHz)
"GH7" (8x A57)

They've been busy the last 6 months, that's for sure.

The thing is, I don't know if the 5433 that was benchmarked is HelsinkiPrime or HelsinkiPro. They don't name the chips by their model numbers in the source from where I gathered that from.
 
The 5433 T760 frequency should be either 650 or 695MHz; no idea about cluster count though. Should be MP6 otherwise it might end slower than in former Exynos SoCs.
 
Nebuchadnezzar said:
Exynos 3250 (2x A7 1GHz Mali 400MP2 133MHz)
Exynos 3470 (4x A7 1.4GHz, Mali 400MP4 450MHz)
Exynos 4415 (4x A9 1.4GHz, Mali 400MP4 533MHz, integrated LTE modem)
Exynos 5250 (2x4 A15/A7, T624 600MHz)
Exynos 5412 (4x4 A15/A7, T624 (??) 600MHz)
Exynos 5422 (4x4 A15/A7, T628MP6 533MHz)
Exynos 5430 (4x4 A15/A7, T760 600MHz)
Exynos 5433 (4x4 A15/A7, T760 ? MHz)
"GH7" (8x A57)

.
Click to expand...

That seems like a crazy amount of chips. Clearly samsung thinks they have the volumes for all of these chips to make them viable, but they don't have great history to date in doing so.
 
Samsung has a ton of fab capacity to fill if Apple's pulling out even partially. Aggressively making more Exynos chips for low and mid-range is pretty expected. They're also filling a bit of a void left by ST-Ericsson and TI no longer making mobile SoCs, although that could also be satisfied by MediaTek.

I also had no idea they had a part with an integrated LTE modem, that seems like a big deal..
 
Exophase said:
Samsung has a ton of fab capacity to fill if Apple's pulling out even partially. Aggressively making more Exynos chips for low and mid-range is pretty expected. They're also filling a bit of a void left by ST-Ericsson and TI no longer making mobile SoCs, although that could also be satisfied by MediaTek.

I also had no idea they had a part with an integrated LTE modem, that seems like a big deal..
Click to expand...
Relatively old news: http://www.koreaherald.com/common_prog/newsprint.php?ud=20131110000061&dt=2

The 4415 is found in some mid-range Korean phones. It's their own SS200 modem in that part. And they seem to want to keep it mid-range: http://www.slideshare.net/fullscree...si-namsung-woo-samsung-system-lsi-business/29

Also, I just noticed in those slides that they say the 5412 was the first WidCon chip. So so much for that mystery model.
 
It's Samsung we're talking about. They need chips for everything: phones, tablets, watches, glasses, chromebooks, TVs, washing machines, ships, construction equipment, smart toilet paper… I'm not sure Exynos can really fit into all of that but there should be no shortage of possible applications.
 
You must log in or register to reply here.

Similar threads

F
Exynos 5450
2 3
Replies
43
Views
14K
Lazy8s
L
D
Creative ZMS-20 and ZMS-40
2 3 4
Replies
75
Views
24K
darkblu
D
M
Samsung Orion SoC - dual-core A9 + "5 times the 3D graphics performance"
16 17 18
Replies
358
Views
113K
Nebuchadnezzar
N
A
Holy Wall of Text: My impressions post Playstation 4 (PS4) unveil
2 3 4
Replies
70
Views
20K
(((interference)))
I
Back
Top