Sempron 2650 AM1 less-obsolete than A10-6800K FM2...

[coercitiv]

but browsing the internet and videos are a breeze

That is just a bad joke.

 

[Insert_Nickname]

while A10-6800K is completely abandoned and obsolete.

Obsolete? Works fine here with Windows 10 Creators Update, and will stay in service for the very foreseeable future after those stunts they've pulled with UHD blurays and Netflix 4K streaming DRM...

 

[waltchan]

Obsolete? Works fine here with Windows 10 Creators Update, and will stay in service for the very foreseeable future after those stunts they've pulled with UHD blurays and Netflix 4K streaming DRM...

According to AMD's driver page, A10-6800K doesn't qualify for newest Crimson ReLive, unless you find a way to hack it (I like to hear it)? Sure, you can install Windows 10 Creators Update, but what AMD graphics driver you're using now?

 

[NTMBK]

Okay, I admit the installation updates were painful to install, but browsing the internet and videos are a breeze. I don't think it's fair for AMD to allow E1-2100 to get all the newest Crimson software, while A10-6800K is completely abandoned and obsolete.

The 6800K uses an older GPU architecture, we've been over this...

 

[Insert_Nickname]

According to AMD's driver page, A10-6800K doesn't qualify for newest Crimson ReLive, unless you find a way to hack it (I like to hear it)? Sure, you can install Windows 10 Creators Update, but what AMD graphics driver you're using now?

Standard issue Windows provided one. I believe it is version 15.7.1. It supports WDDM 1.3 at least.

You're right about Crimson support, but in practice you're only missing out on the new control panel. For a set-and-forget system, the old one works just fine. No new features are added for VLIW5/4 GPUs anyway, and I don't think there is much performance optimisation that makes sense anyway. Slow +5-10% is still slow.

 

[Spjut]

I fixed Windows 10 on a PC with an A8 3870k and was surprised to see that the driver is old compared to the last driver for HD 5000 and 6000 series. It reports WDDM 1.2 in DXdiag, but it's Terascale 2, just as HD 5000, which reports WDDM 1.3 and was officially supported until january 2016.

 

[SPBHM]

I fixed Windows 10 on a PC with an A8 3870k and was surprised to see that the driver is old compared to the last driver for HD 5000 and 6000 series. It reports WDDM 1.2 in DXdiag, but it's Terascale 2, just as HD 5000, which reports WDDM 1.3 and was officially supported until january 2016.

the last WHQL driver for the HD5000 is 15.7
the support ended still in 2015, the last driver under normal support was 15.11.1 beta, after that it received 2 beta drivers as a "legacy card", they added the new control panel, but missing features in relation to the GCN version, and also with no GPU OpenCL, other that that it seems to be basically the same driver as the 2015 drivers.

Llano missing WDDM 1.3 is really odd, but not surprising, AMD typically supported poorly their previous series, GCN is the exception.
Fermi is still being supported and I think received WDDM 2.2 drivers recently.

the older DX11 Radeons are showing several glitches in newer titles due to being abandoned.
http://forums.guru3d.com/showthread.php?t=413434

but back to the OP, I much rather use Trinity/Richland with drivers from 2015 than something as terrible as that with updated drivers,
yet it's a shame that Haswell IGP still gets new drivers all the time, but the much faster APU AMD was selling at the same time does not for over a year now.

 

[waltchan]

The 6800K uses an older GPU architecture, we've been over this...

Yes, I know. AMD should put restriction updates based on price and speed.

 

[DrMrLordX]

Based on price? That's silly.

 

[waltchan]

Based on price? That's silly.

I think it makes more sense. This pressures people to get off their very slow one for faster one. Who uses E1-2100 for daily use, AMD should restrict this model.

 

[Shivansps]

the last WHQL driver for the HD5000 is 15.7
the support ended still in 2015, the last driver under normal support was 15.11.1 beta, after that it received 2 beta drivers as a "legacy card", they added the new control panel, but missing features in relation to the GCN version, and also with no GPU OpenCL, other that that it seems to be basically the same driver as the 2015 drivers.

Llano missing WDDM 1.3 is really odd, but not surprising, AMD typically supported poorly their previous series, GCN is the exception.
Fermi is still being supported and I think received WDDM 2.2 drivers recently.

the older DX11 Radeons are showing several glitches in newer titles due to being abandoned.
http://forums.guru3d.com/showthread.php?t=413434

but back to the OP, I much rather use Trinity/Richland with drivers from 2015 than something as terrible as that with updated drivers,
yet it's a shame that Haswell IGP still gets new drivers all the time, but the much faster APU AMD was selling at the same time does not for over a year now.

Funny thing is that they are still selling Richland as A4-7300.

 

[DrMrLordX]

I think it makes more sense. This pressures people to get off their very slow one for faster one. Who uses E1-2100 for daily use, AMD should restrict this model.

Not that the E1-2100 or A10-6800k factor into AMD's plans moving forward, but . . . fact is, the A10-6800k was a upper-mid-tier product using the old VLIW 4 while the E1-2100 is a bottom-basement product that just happened to come out after AMD had adopted GCN. AMD chooses to support GCN but not VLIW 4. It really isn't that hard to figure out.

Continuing to support VLIW4 with driver updates would contribute to driver bloat (or promote driver bifurcation), while categorically failing to support the E1-2100 simply makes no sense from a technical point of view since the driver already supports other GCN products, all the way back to GCN 1.0 .

The price or relative usefulness of the APUs in question has no bearing on whether or not it makes any sense for AMD to support them. AMD doesn't support VLIW 4 anymore, period. It takes zero effort for them to support the E1-2100 so why not?