AMD Ryzen 5 2400G and Ryzen 3 2200G APUs performance unveiled

Page 25 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

Shivansps

Diamond Member
Sep 11, 2013
3,873
1,527
136
How much ram will these APUs be able to use for graphics?

If they can use up to 4gb, that's a substantial chunk of system ram.

You'd want more than 8gb total ram, to allow the APU to use 4gb.

Well, it has 2GB max of dynamic memory, unless that was changed in RR driver I dont think it can use 4GB.

Memory is another issue here, with 8GB of ram, you can only use 6GB, compared to full 8GB of a dgpu build, if you add 2 sticks of 4 GB, in a cheap MB you will have a problem to upgrade that later on, compared to just add a single 8GB stick with a dgpu.

Anyway i would not worry about the ram, the only 2 games that i know they cant be played with 4GB of ram are Assassin Creed Origins(6GB is the minimum) and 7 days to die(even 8 is not enoght). There is probably a few more indie voxel based that abuses of ram.

And from what i tested, the RX560 4GB performed better than the 1050 2GB when is accompanied with 4GB of RAM.
 
Last edited:

french toast

Senior member
Feb 22, 2017
988
825
136
Yes mining has ruined pc gaming.. memory prices have not helped either, nor the lack of competition from AMD.
I'm actually starting to consider an Xbox one X to tide me over, but I really want a pc build, I'm leaning back towards a 2200g, 1 stick 8gb ddr4 3200, +dgpu...but you prices are crazy right now... I'm not paying more than £160 for a 1050ti..no way.
 

wahdangun

Golden Member
Feb 3, 2011
1,007
148
106
Yeah memory price is the real evil, but somehow it's not showing in news. I mean we don't need dgpu for PC to function but we really need ram to even booting.


If just Xbox one x can be used for mining, then the burden can be spread.
 

neblogai

Member
Oct 29, 2017
144
49
101
Here is how much Wolfenstein2 performance R5 2500U in HP x360 gains when new drivers are force-fed to it: https://www.youtube.com/watch?v=-B6FlUA9PKI . It is of course an SKU very similar to 2200G. Differences are:
1. 2200G CPU does not support SMT, but can be expected to work at constant ~3.5GHz and boosting higher, compared to ~2.0-2.5GHz on a mobile chip.
2. GPU clocks: we see them in ~850MHz range for the 2500U- while 2200G should be overclockable to at least 1300-1400MHz with the stock cooler. That would be ~60% increase.
3. Available memory bandwidth: even with with DDR4 2400 memory installed on x360- we see it working at 1866MHz most of the time, with jumps to 2133MHz. So using supported 2933MHz is also 35-60% increase in bandwidth.
Current x360 reviews show 2500U with old drivers to perform at about 60% of the MX150 in various games. New drivers bring about 20% extra performance in Wolfenstein, and users report other games benefiting from them as well. Higher iGPU and memory clocks for the desktop 2200G should bring ~50% on top of that. So using very approximate numbers- 0.6 x 1.2 x 1.5= 1.08 of MX150 performance for the 2200G, or maybe 20% more if 11CU 2400G and DDR4 3200 memory is used instead.
 

neblogai

Member
Oct 29, 2017
144
49
101
You are joking, right?

AMD has already revealed the performance of an 2400G in various configurations.
MX 150 scores around 3700 in Fire Strike.

1. 3DMark scores have been very unreliable for estimating APU performance, so I'd rather not rely on them.
2. I do remember that 2000G series APUs are memory bandwidth limited, which brings the question of game settings: I based that estimation of mine on 720p low/medium- but I do think GT1030/RX550 will probably be faster at 1080p. My opinion is: 2000G APUs can be faster, but will need to be used at game settings optimized for them to avoid their bottleneck.
 
Reactions: french toast

LTC8K6

Lifer
Mar 10, 2004
28,520
1,575
126
I think they will turn out to be too thermally and bandwidth limited to perform like we might think they should on paper. The memory bandwidth is only about 1/3 that of a 1050. Even the lowly 64bit 1030 has slightly more memory bandwidth. They will still be the best integrated solution, though.
 

raghu78

Diamond Member
Aug 23, 2012
4,093
1,475
136
Here is how much Wolfenstein2 performance R5 2500U in HP x360 gains when new drivers are force-fed to it: https://www.youtube.com/watch?v=-B6FlUA9PKI .

Thats quite a bit of improvement. I am waiting to see how the Ryzen 65w APUs perform with DDR4 3200 in games.

It is of course an SKU very similar to 2200G. Differences are:
1. 2200G CPU does not support SMT, but can be expected to work at constant ~3.5GHz and boosting higher, compared to ~2.0-2.5GHz on a mobile chip.
2. GPU clocks: we see them in ~850MHz range for the 2500U- while 2200G should be overclockable to at least 1300-1400MHz with the stock cooler. That would be ~60% increase.
3. Available memory bandwidth: even with with DDR4 2400 memory installed on x360- we see it working at 1866MHz most of the time, with jumps to 2133MHz. So using supported 2933MHz is also 35-60% increase in bandwidth.
Current x360 reviews show 2500U with old drivers to perform at about 60% of the MX150 in various games. New drivers bring about 20% extra performance in Wolfenstein, and users report other games benefiting from them as well. Higher iGPU and memory clocks for the desktop 2200G should bring ~50% on top of that. So using very approximate numbers- 0.6 x 1.2 x 1.5= 1.08 of MX150 performance for the 2200G, or maybe 20% more if 11CU 2400G and DDR4 3200 memory is used instead.

Performance in games might get very close to GT 1030 but we need to wait and see with proper testing how the bandwidth bottlenecks affect perf. Remember APUs have to share 50GB /s bandwidth (2ch DDR4 3200) between CPU and GPU. So its not going to be easy to match or exceed GT 1030. I am excited about the Ryzen APUs and what they could achieve a couple of generations down the road at 7nm with a single stack of HBM2.
 

PeterScott

Platinum Member
Jul 7, 2017
2,605
1,540
136
Performance in games might get very close to GT 1030 but we need to wait and see with proper testing how the bandwidth bottlenecks affect perf. Remember APUs have to share 50GB /s bandwidth (2ch DDR4 3200) between CPU and GPU. So its not going to be easy to match or exceed GT 1030. I am excited about the Ryzen APUs and what they could achieve a couple of generations down the road at 7nm with a single stack of HBM2.

It's nice that it might get close to matching a GT 1030, if a GT 1030 was a GPU card you would actually consider buying.

But I would never consider buying GT 1030 card.

But I would still love to see this in a Mac Mini where a dGPU hasn't been an option in years.
 

whm1974

Diamond Member
Jul 24, 2016
9,460
1,570
96
It's nice that it might get close to matching a GT 1030, if a GT 1030 was a GPU card you would actually consider buying.

But I would never consider buying GT 1030 card.

But I would still love to see this in a Mac Mini where a dGPU hasn't been an option in years.
Depending on the use case I would. But yeah the GT 1030 GPU wouldn't normally be my first choice either. And I agree that 2200G and 2400G will be good thing for SFF and AIO systems.
 

raghu78

Diamond Member
Aug 23, 2012
4,093
1,475
136
It's nice that it might get close to matching a GT 1030, if a GT 1030 was a GPU card you would actually consider buying.

But I would never consider buying GT 1030 card.

But I would still love to see this in a Mac Mini where a dGPU hasn't been an option in years.
The fact that Nvidia is selling GT 1030 for desktops and MX150 for notebooks means there is a market for that kind of price and perf. It does not matter what you or I would consider buying. Frankly i would not buy a GPU less than Rx 570 level perf and with any less than 4GB VRAM. But i would also not pay more than USD 180 which in these crypto mining inflated dGPU channel pricing is not happening.

Sent from my SM-G935V using Tapatalk
 

Shivansps

Diamond Member
Sep 11, 2013
3,873
1,527
136
for the 10th time, AMD avoided to show a direct game comparison with GT1030, this is something they always did in the past, if the numbers favor them, there is really no much to add, to me it cant match it and maybe it cant get close enoght either. What would be sad because is a big regresion in IGP vs dGPU perf ratio.

And those firestrike numbers(what is a best case escenario) are not that good vs GT1030, and to counter that AMD decided to add two overclocked results what to me it means its bad.
 
Last edited:

Gideon

Golden Member
Nov 27, 2007
1,712
3,931
136
I just noticed something interesting with AMD 25x20 graph on this article: https://www.anandtech.com/show/1231...n-exclusive-interview-with-dr-lisa-su-amd-ceo


I think this is because the actual 1-25x "line" is logarithmic, not linear. Raven Ridge (5.86x) on this chart is barely above Carrizo (3.5x), yet it's nearly 50% more power efficient. On a linear chart, it should be about 50% of the distance between Carrizo and Kaveri (as it is on your chart).

It also makes sense to portrait it this way. A 7nm APU will, in all likelygood, be at least 2x more efficient than Raven Ridge (given the process advantages alone) and You would want to plot reasonably (otherwise it would go off the slide quickly). Therefore a non-linear "Required Trend" line was chosen.
 

AtenRa

Lifer
Feb 2, 2009
14,003
3,361
136
Last edited:

Shivansps

Diamond Member
Sep 11, 2013
3,873
1,527
136
The fact that Nvidia is selling GT 1030 for desktops and MX150 for notebooks means there is a market for that kind of price and perf. It does not matter what you or I would consider buying. Frankly i would not buy a GPU less than Rx 570 level perf and with any less than 4GB VRAM. But i would also not pay more than USD 180 which in these crypto mining inflated dGPU channel pricing is not happening.

Sent from my SM-G935V using Tapatalk

Of course there is a market, and the GT1030 is almost a GTX750TI, its really impresive what Nvidia archived with the GT1030, 2 years ago you could never got anywhere near this kind of performance for just $70-80, then again this is entry level a gpu for notebooks. IF the MX150 did not existed, they would have rebranded a Maxwell or something.

Also is not the same thing a dGPU that you can put on a pc you already have, than selling a $170 APU with high speed ram and then ask it to OC! And it may not be enoght.
 

french toast

Senior member
Feb 22, 2017
988
825
136
It's nice that it might get close to matching a GT 1030, if a GT 1030 was a GPU card you would actually consider buying.

But I would never consider buying GT 1030 card.

But I would still love to see this in a Mac Mini where a dGPU hasn't been an option in years.
I don't understand what exactly you are trying to say....which gpu would you prefer then for 25$ or 80$??..(2200g ,2400g and pentium are 99$/180$/75$ respectably).. remember we are talking about sub 200$ CPU & GPU combo for budget gaming..there are not many options on the table.

Depending on the use case I would. But yeah the GT 1030 GPU wouldn't normally be my first choice either. And I agree that 2200G and 2400G will be good thing for SFF and AIO systems.
Agreed.

for the 10th time, AMD avoided to show a direct game comparison with GT1030, this is something they always did in the past, if the numbers favor them, there is really no much to add, to me it cant match it and maybe it cant get close enoght either. What would be sad because is a big regresion in IGP vs dGPU perf ratio.

And those firestrike numbers(what is a best case escenario) are not that good vs GT1030, and to counter that AMD decided to add two overclocked results what to me it means its bad.
Why should they?? AMDs product is an Apu, it should be compared to other APUs, which it is now the market leader (in regards to graphics mated to a decent cpu)...why would you expect a company to advertise their products in a worse situation than they have to?
In the price range to compete..you literally have to buy the crappiest cheapest bespoke descrete options from 2 leading competitor's...even then whilst you may get higher gaming performance...you lose out on CPU performance...on future upgrades to 7nm...on chassis space and likely power consumption?..small extra effort/inconvenience...all these things add up for some people/OEMs.

Not to mention R3 1200 Is now about the same price as the Pentiums..before 2200g launches...what stops you from mating a dgpu to either the 2200g or the cheaper R3 1200??...

Usually, all- in -one tech products are more expensive..as you pay for the miniaturisation and technology to combine it into one no nonsense package, before now there was large sacrifices on either CPU or GPU (or both)...this is the first time I can remember that you get an APU with almost no serious compromises for its segment...no obvious downsides....each part of the Apu offers performance expected in its price and market segment...whilst being fully upgradeable into future products.
It's a win win.

Besides, not everyone is a gamer...office computers are going to love these little APUs.
 
Last edited:

LTC8K6

Lifer
Mar 10, 2004
28,520
1,575
126
Office computers need the barest minimum of graphics capabilities. Even a GT710 is plenty.
The GPU in the 2200G/2400G is way more than an office computer needs.
 
Reactions: french toast

IRobot23

Senior member
Jul 3, 2017
601
183
76
You are joking, right?

AMD has already revealed the performance of an 2400G in various configurations.



MX 150 scores around 3700 in Fire Strike.

Nope it does not.
MX 150 AVG is around 3Kp, GT 1030 around 3350p based on notebookckeck.net .
 
sale-70-410-exam    | Exam-200-125-pdf    | we-sale-70-410-exam    | hot-sale-70-410-exam    | Latest-exam-700-603-Dumps    | Dumps-98-363-exams-date    | Certs-200-125-date    | Dumps-300-075-exams-date    | hot-sale-book-C8010-726-book    | Hot-Sale-200-310-Exam    | Exam-Description-200-310-dumps?    | hot-sale-book-200-125-book    | Latest-Updated-300-209-Exam    | Dumps-210-260-exams-date    | Download-200-125-Exam-PDF    | Exam-Description-300-101-dumps    | Certs-300-101-date    | Hot-Sale-300-075-Exam    | Latest-exam-200-125-Dumps    | Exam-Description-200-125-dumps    | Latest-Updated-300-075-Exam    | hot-sale-book-210-260-book    | Dumps-200-901-exams-date    | Certs-200-901-date    | Latest-exam-1Z0-062-Dumps    | Hot-Sale-1Z0-062-Exam    | Certs-CSSLP-date    | 100%-Pass-70-383-Exams    | Latest-JN0-360-real-exam-questions    | 100%-Pass-4A0-100-Real-Exam-Questions    | Dumps-300-135-exams-date    | Passed-200-105-Tech-Exams    | Latest-Updated-200-310-Exam    | Download-300-070-Exam-PDF    | Hot-Sale-JN0-360-Exam    | 100%-Pass-JN0-360-Exams    | 100%-Pass-JN0-360-Real-Exam-Questions    | Dumps-JN0-360-exams-date    | Exam-Description-1Z0-876-dumps    | Latest-exam-1Z0-876-Dumps    | Dumps-HPE0-Y53-exams-date    | 2017-Latest-HPE0-Y53-Exam    | 100%-Pass-HPE0-Y53-Real-Exam-Questions    | Pass-4A0-100-Exam    | Latest-4A0-100-Questions    | Dumps-98-365-exams-date    | 2017-Latest-98-365-Exam    | 100%-Pass-VCS-254-Exams    | 2017-Latest-VCS-273-Exam    | Dumps-200-355-exams-date    | 2017-Latest-300-320-Exam    | Pass-300-101-Exam    | 100%-Pass-300-115-Exams    |
http://www.portvapes.co.uk/    | http://www.portvapes.co.uk/    |