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

Page 81 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

SPBHM

Diamond Member
Sep 12, 2012
5,056
409
126
AMD isn't a charity. Not to mention these are unlikely to go into a board that supports overclocking. There is no reason for such a cheap part to be unlocked.

they are charging money for it so sure, not a charity, but I don't think it costs anything extra for AMD to leave the multiplier unlocked,

the cheaper B350 boards are perfectly fine and priced well enough for this kind of CPU
OC would be another solid reason to take one of these over the Intel stuff, oh well.
 

VirtualLarry

No Lifer
Aug 25, 2001
56,229
9,990
126
Two thoughts,

1) I thought that AMD said before that the AM4 platform would be unlocked, and
2) What does the boxed cooler for these CPUs look like? Does it require removing the heatsink mounting tabs, like the Ryzen heatsinks do? I would personally think, that it would be like the Bristol Ridge heatsinks, and the Athlon II etc. heatsinks before it, the standard aluminum skived block, with the standard AMD heatsink mounting tab mechanism. I guess we'll see, soon enough.

When do these go on sale?
 

f2bnp

Member
May 25, 2015
156
93
101
Sweet little part, makes for a great entry level CPU and it also has a basic iGPU. I see it as a great replacement to the Stoney Ridge parts and it serves a great role as a daily driver for most people that want a computer that's fast enough for office work and the such. Using a CPU like this you can keep the costs down by using a cheap CPU + motherboard + RAM (no need to aim for dual channel here), while also investing in parts that matter far more for a smooth user experience such as a good SSD.
 
  • Like
Reactions: Insert_Nickname

DrMrLordX

Lifer
Apr 27, 2000
21,583
10,785
136
Guess it's worth reiterating that Intel OEMs don't pay anywhere near MSRP prices. Don't know how AMD does it but wouldn't be surprised if it is the same way. Either way you can't really use MSRP for what a Dell would pay.

Unless Intel is giving bigger breaks to the OEM than AMD, you can approximate the relative cost to the OEM based on typical retail profit percentages. Usually retail prices are 40% profit, though how the 40% breaks down between reseller, distributor/supplier, and manufacturer is not always uniform.

Intel does strive for higher margins than AMD, so it is possible that the G5500's MSRP is closer to 50-60% profit. But we don't know that for sure. The only knowledge we have are relative prices based on MSRPs and/or obvious street pricing.
 

Shivansps

Diamond Member
Sep 11, 2013
3,835
1,514
136
There is really no reason for these Athlons to be locked, UNLESS, there is a 4/4 Athlon, and i think the 240GE is 4/4, probably a desktop version of the 2300U.
 

jpiniero

Lifer
Oct 1, 2010
14,511
5,159
136
There is really no reason for these Athlons to be locked, UNLESS, there is a 4/4 Athlon, and i think the 240GE is 4/4, probably a desktop version of the 2300U.

If this is a salvaged QC Raven Ridge die, AMD is probably "losing money" on the 200GE because of how cheap it is relative to the die size. It's better than the trash can of course but you can't make it that appealing.
 

JustMe21

Senior member
Sep 8, 2011
324
49
91
Does anyone know if it's possible to underclock the Ryzen 5 2400g so it runs around 2 GHz, similar to the V1605B, but with more Compute units? If it is possible, it's cheaper to buy the 2400G and underclock it than to try and find an embedded V1605B motherboard.
 

ZGR

Platinum Member
Oct 26, 2012
2,052
656
136
Does anyone know if it's possible to underclock the Ryzen 5 2400g so it runs around 2 GHz, similar to the V1605B, but with more Compute units? If it is possible, it's cheaper to buy the 2400G and underclock it than to try and find an embedded V1605B motherboard.

Easily done. Might as well try to undervolt it too.
 

helpful55

Junior Member
Oct 21, 2014
6
1
66
Thanks you all for adding comments to the ACPI situation.
Here are some updates and comments:

  • Slipstreaming EHCI / XCHI USB drivers is intended to allow the use of a USB keyboard/mouse in order to install win7, but it is useful only if there is not a PS/2 port or PS/2 keyboard available. With the exception of the Asus ROG Strix B350-F, every other AM4 board I have used has a PS/2 port, and that includes the upper higher end like the Asus Crosshair hero VI, Gigabyte Aorus GA-AX370 Gaming K7 and AsRock X370 Fatal1ty Gaming Pro. All of them have a PS/2 port.
  • During a win7 fresh install, regardless if slipstreamed usb drivers or PS/2 keyboard use, at some point, install will crash throwing the ACPI error.
  • For sanity, I tried using an RX 550 with the 2400G in the prime X370 pro, disabling the iGPU. Still crashed trying to boot into windows 7.
  • I put a Ryzen 5 1600X again in the AB350 Gaming 3, tried once more with the SSD that had the win7 image restored. It booted windows 7 fine. F20 BIOS
  • I did the same with the prime x370 pro, I installed a Ryzen 5 1600. Win7 booted fine. 3803 BIOS. This test, along with the 1600X in the AB350 gaming 3 suggests that the BIOSes are fine, the ACPI situation is at the hardware level inside the Ryzen G APUs.
  • These UEFI BIOSes contain AGESA 1.0.0.0a. I wonder if the BIOSes wth AGESA 1.0.7.2 would be able to recognize the Ryzen APU, AND be able to boot win7. I will try it.
  • Surprisingly, win 7 runs in raven ridge mobile. The HP envy x360 with Ryzen 5 2500u can boot windows 7

The ACPI situation might get fixed with future BIOS updates, it was similar in many OEM PCs during Win8 launch, systems crashing if trying to install/run win7, and eventual BIOS updates allowed win7 to run.
In the meantime, no windows 7 on Ryzen desktop APUs at all.
This is a roadblock for some of us. I am a system builder, and while I try to steer every customer toward windows 10, win7 still has a loyal following. To make it worse, the win7 loyalty is higher at the lower price points, so the crowd that would benefit the most form Ryzen APUs are the ones that want windows 7... so it stinks! On the upside, this might be what I needed to have everyone running win 10.

On a subjective note, the 2400G feels every bit as snappy as a 1500X.
Did you ever try the other bioses? I am still hoping to do W7 on a 2400G. And as far as W10 or maybe any ms product, I wish we could add the features everyone likes and leave out all the stuff ms likes to linux and we could have real working pc's. I am still hoping to do W7 on a 2400G. Thanks.
 

alexruiz

Platinum Member
Sep 21, 2001
2,836
556
126
Did you ever try the other bioses? I am still hoping to do W7 on a 2400G. And as far as W10 or maybe any ms product, I wish we could add the features everyone likes and leave out all the stuff ms likes to linux and we could have real working pc's. I am still hoping to do W7 on a 2400G. Thanks.

I didn't try the older BIOSes.
I might try one with AGESA PinacclePi 1.0.0.6, but honestly I lost interest in Win7 and APUs.
 

helpful55

Junior Member
Oct 21, 2014
6
1
66
Yes really. For myself I rather not even use Win10 unless I have to, Hell I had to install ClassicShell to make Windows 8 usable.
It looks like my plan is to use 8.1 maybe after 7 support ends. I would like to stay with the great 7, but 8.1 seems like it will be tolerable. Classic shell makes most things awfully similar. No interest in 10 unless I have to. Don't listen to lazy 10'ers.
 

VirtualLarry

No Lifer
Aug 25, 2001
56,229
9,990
126
Yeah, I saw that Reddit. Still doesn't work on my ASRock AB350M Pro4 boards, with AGESA 1.0.0.6.

Maybe it will "run", if you install it on a DIFFERENT AM4 rig, and then move the SSD over. But it won't INSTALL FRESH on a 2200G, even WITH AGESA 1.0.0.6.
 

Cormy1

Junior Member
Jan 7, 2020
2
1
41
Thanks you all for adding comments to the ACPI situation.
Here are some updates and comments:

  • Slipstreaming EHCI / XCHI USB drivers is intended to allow the use of a USB keyboard/mouse in order to install win7, but it is useful only if there is not a PS/2 port or PS/2 keyboard available. With the exception of the Asus ROG Strix B350-F, every other AM4 board I have used has a PS/2 port, and that includes the upper higher end like the Asus Crosshair hero VI, Gigabyte Aorus GA-AX370 Gaming K7 and AsRock X370 Fatal1ty Gaming Pro. All of them have a PS/2 port.
  • During a win7 fresh install, regardless if slipstreamed usb drivers or PS/2 keyboard use, at some point, install will crash throwing the ACPI error.
  • For sanity, I tried using an RX 550 with the 2400G in the prime X370 pro, disabling the iGPU. Still crashed trying to boot into windows 7.
  • I put a Ryzen 5 1600X again in the AB350 Gaming 3, tried once more with the SSD that had the win7 image restored. It booted windows 7 fine. F20 BIOS
  • I did the same with the prime x370 pro, I installed a Ryzen 5 1600. Win7 booted fine. 3803 BIOS. This test, along with the 1600X in the AB350 gaming 3 suggests that the BIOSes are fine, the ACPI situation is at the hardware level inside the Ryzen G APUs.
  • These UEFI BIOSes contain AGESA 1.0.0.0a. I wonder if the BIOSes wth AGESA 1.0.7.2 would be able to recognize the Ryzen APU, AND be able to boot win7. I will try it.
  • Surprisingly, win 7 runs in raven ridge mobile. The HP envy x360 with Ryzen 5 2500u can boot windows 7

The ACPI situation might get fixed with future BIOS updates, it was similar in many OEM PCs during Win8 launch, systems crashing if trying to install/run win7, and eventual BIOS updates allowed win7 to run.
In the meantime, no windows 7 on Ryzen desktop APUs at all.
This is a roadblock for some of us. I am a system builder, and while I try to steer every customer toward windows 10, win7 still has a loyal following. To make it worse, the win7 loyalty is higher at the lower price points, so the crowd that would benefit the most form Ryzen APUs are the ones that want windows 7... so it stinks! On the upside, this might be what I needed to have everyone running win 10.

On a subjective note, the 2400G feels every bit as snappy as a 1500X.

I just want to note that I have a working rig with a 2400G and Windows 7 installation.
Do not ask me how to reproduce it, I don't know how I managed to do it and I can't reproduce it.
I can describe how I got here though, but I don't think it will help anyone.
I had a build using an HD6850 GPU from AMD, this one in particular: https://www.techpowerup.com/gpu-specs/sapphire-hd-6850-1440sp-edition.b1892
I was running on a Phenom II X4 970 with an M4A89GTD Motherboard. https://www.asus.com/Motherboards/M4A89GTD_PROUSB3/
I had replaced the HD 6850 by a GTX 750TI.
So assume I had all the drivers for that stuff already installed on this drive.
I gut the PC, replacing the PSU, Mobo, CPU and RAM. I flash the BIOS to the latest, going from F20 to F31 to F50A on https://www.gigabyte.com/Motherboard/GA-AB350-Gaming-rev-1x
At the beginning I was having trouble getting visuals, so I took out the GTX 750TI. I'm not sure at what point I had reinserted it, but I had gone into this knowing that Win7 was never going to work with the iGPU so I believe I reinserted it and got it working before I tried booting into my drives.
I re-insert my drives. I had 2 drives, one was the old HDD that has over 3GB of drivers on it, and an SSD with about 1.25GB of drivers. Both were installed on the same system and both were functioning fine before. I transferred them to the new rig and tried to boot into them for fun. Naturally, it didn't work. I also had a recovery disc which I attempted to use, which also didn't work. I'm not exactly sure which steps were taken and in what order.
In any case, I had given up and THOUGHT I had chosen to boot into my Windows 10 installation media, which was a USB.
To my surprise, when I rebooted it was the Windows 7 logo I was looking at. It sat there for a long while and I just let it do its thing.
Eventually it got to the User Login page, lo and behold my users were there. The resolution was wrong, 480p like I was in safe mode or some recovery environment, but it worked. From there, Windows started installing a bunch of drivers on its own, I don't know where it was getting the drivers from but I have to assume it was from either my Windows 10 installation disc, or from the motherboard's installation disc from Gigabyte.
Sufficed to say, as soon as I was in I started installing everything I believed I needed, chipsets for Windows 7 from Gigabyte's support page for my motherboard, and what I believed I required from the motherboard installation disc.
From that point on it just worked, and I got the proper resolution back.
I haven't been able to reproduce this on my second drive, only the one that unexpectedly booted when I didn't want it to.
I don't know where to begin with figuring all this out.
 

Attachments

  • Win7 2400G.png
    Win7 2400G.png
    225.2 KB · Views: 16
  • DSCN2119.jpg
    DSCN2119.jpg
    1.1 MB · Views: 15

Cormy1

Junior Member
Jan 7, 2020
2
1
41
Update on this, someone's modding on Win-Raid to get compatibility on up to 4000 series APUs. Seems they got it working a month before I did, on purpose rather than accidentally...
 
  • Like
Reactions: drSeehas