AMD's Tonga - R9 285 (Specs) and R9 285X (Partial Specs)

[JDG1980]

I'm starting to wonder if there's something more going on here, like legal threats or marketing threats. (If I were in AMD marketing... well, first of all I'd die of shame, but secondly I'd be livid that the company is paying Anandtech for a special "AMD Center" sponsorship yet can't even get timely product reviews.)

 

[wand3r3r]

If it happens again then it may be worth bringing up, however (imo) considering that the sites operations are going through a major revamp it's not surprising they are scrambling a bit. I doubt there's any major conspiracy going on here.

If I have to guess I'd just say the major GPU reviewer (Ryan) simply got the workload of Anand + Ryan and is swamped atm. If there are other gpu reviews being delayed in the future, well there's no sense speculating now.

 

[yannigr]

The newest article comes top. The Xeon article was not as new as the Lumia article. Also, the Lumia article was 9 pages, Xeon article was 22- and finishes with "Conclusions so Far...", implying that there is more to come. Don't be ridiculous.

Now that thing in the end about being ridiculous, you really think that makes ME look ridiculous and not the one who writes it?

You can express your opinion and still be polite. It's not difficult. Trust me.

 

[realibrad]

Now that thing in the end about being ridiculous, you really think that makes ME look ridiculous and not the one who writes it?

You can express your opinion and still be polite. It's not difficult. Trust me.

Yes, your comment was ridiculous, because its obvious. The fact that you were complaining about an article is no longer first, because another newer article came out is absurd.

 

[yannigr]

Yes, your comment was ridiculous, because its obvious. The fact that you were complaining about an article is no longer first, because another newer article came out is absurd.

No, your wrong conclusion is ridiculous. In the future try to distinguish between an observation and a complain. And next time you call someone ridiculous, try doing it in front of a mirror not a monitor(except if you are using a camera of course).

 

[3DVagabond]

If Ryan hires someone else to be the video reviewer it might be a good opportunity to review the process and make some improvements. No offense to Ryan or the current system, but it would be an opportunity to look at possibly improving things. It's hard to change review processes. You have people who want every new technique added but the need for consistency are at odds with that. With a new reviewer though, consistency would be near impossible anyway.

 

[Enigmoid]

AT needs to fix their GPU review system too in that power should be isolated at the card level, not simply system power (includes CPU).

 

[el etro]

Hey AT, come on with the R9 285 review!!

 

[TheRealSintel]

Don't bother waiting for it anymore, if it hasn't been done now it never will. The tech report has a good take on it.

Remember the Nexus 10 review?
http://forums.anandtech.com/showthread.php?t=2287750

 

[realibrad]

No, your wrong conclusion is ridiculous. In the future try to distinguish between an observation and a complain. And next time you call someone ridiculous, try doing it in front of a mirror not a monitor(except if you are using a camera of course).

So an observation is this...

The day is warmer than the night.

A complaint is this...

I dont like that the day is warmer than the night.

What you pointed out was not a neutral statement, but something you did not like, aka complaint. It does not appear that you randomly point out things, so given the context, the reason for your comment is a complaint.

Also, nice job on the hypocrisy.

 

[yannigr]

So an observation is this...

The day is warmer than the night.

A complaint is this...

I dont like that the day is warmer than the night.

What you pointed out was not a neutral statement, but something you did not like, aka complaint. It does not appear that you randomly point out things, so given the context, the reason for your comment is a complaint.

Also, nice job on the hypocrisy.

Are 14 years old or just a troll? And stop quoting me everytime you talk to the mirror.

 

[tviceman]

AT needs to fix their GPU review system too in that power should be isolated at the card level, not simply system power (includes CPU).

There are merits to measuring power either way.

 

[Gloomy]

There are merits to measuring power either way.

I can't think of any.

 

[96Firebird]

I can't think of any.

Total system is what most people will use to judge how much of a PSU they need. If a certain video card tends to make the CPU run with more juice (for whatever reason), the owner may need a more powerful PSU.

Card power usage is good for comparing card efficiency vs other cards. Also, can usually be used to figure out how efficient a GPU is (and, in turn, architecture). So long as the temps are in check.

 

[Termie]

There are merits to measuring power either way.

I can't think of any.

More powerful GPUs will drive the rest of the system harder, leading to higher power draw. It all comes down to whether you care what your GPU is drawing in isolation, or how much power your entire system is using.

The latter is much more important if you're trying to equip your system with the appropriate power supply, as well as to calculate electricity costs. The former is only important in theoretical power/Watt/mm2 discussions.

 

[yannigr]

Review posted. Great!

 

[ShintaiDK]

And 279$ again posted for the 280, while its 249$

http://www.anandtech.com/show/8019/amd-cuts-radeon-r9-280-to-249

And power consumption numbers. Basicly same as 280 tho, even with 60W less TDP. Seems to be a habbit :/

The memory compression shows its clear benefit tho. And why you cant compare VRAM as such or memory bandwidth.

And good to see AMD now finally got a card that can actually decode 4K.

It also raises the ugly Mantle flag again.

Because Tonga is based on a new GPU – and a newer version of GCN no less – the developers of Thief and Battlefield 4 have not had the opportunity to optimize their games for Tonga products. If you have ever used some of the lower end GCN products (e.g. Cape Verde) then you’ve seen first-hand that these games already are hit & miss depending on the GPU in use, so Tonga is an extension to that limitation.

On the one hand Mantle is up and running; both Thief and Battlefield 4’s Mantle rendering paths work on R9 285 despite neither game having seen the GPU before, and as far as we can tell there are no immediate rendering errors. However the fact that Mantle performance has significantly regressed and at this point is below Direct3D performance is not what we’d like to see

Thats just an utter fail.

 

[96Firebird]

And 279$ again posted for the 280, while its 249$

To be fair, the chart row is labeled as "Launch Price", not current price or MSRP. Why that is, I have no idea.

 

[Enigmoid]

Mantle has advantages but it is even more dependent on driver support. The game dev has a lot of control but what if 3+ years down the line you launch a new GPU architecture? The game maker isn't going to modify the game after than long so it falls on your shoulders to try and fix the bugs, which on such a low level architecture may be significantly harder.

This also shows the dependence of mantle on fine tuning the design to the game; without that fine tuning you may have problems.

It looks like the 760 and 770 have dropped almost perfectly in price. Cheapest 760 and 770 are $205 (after rebate) and $280 respectively.

http://www.newegg.com/Product/Produc...D=3938566&SID=

http://www.newegg.com/Product/Produc...D=3938566&SID=

 

[monstercameron]

^an utter fail that can be solved by a single patch.

this is the perfect exercise to see what happens when mantle is ran on a newer/different gpu isa.

 

[ShintaiDK]

^an utter fail that can be solved by a single patch.

this is the perfect exercise to see what happens when mantle is ran on a newer/different gpu isa.

And who is going to pay for that patch down the road?

Lets be honest, we all know what happens. No patch unless someone sponsor it.

 

[monstercameron]

And who is going to pay for that patch down the road?

Lets be honest, we all know what happens. No patch unless someone sponsor it.

No, we all dont know that. So you are claiming that amd, nvidia and intel pay every game dev to patch their games? This is so silly, I can't even...

 

[tviceman]

I can't think of any.

Think harder.

 

[ShintaiDK]

No, we all dont know that. So you are claiming that amd, nvidia and intel pay every game dev to patch their games? This is so silly, I can't even...

We are talking about a patch to up the performance for an API for a single GPU that already works.

Its not a game patch. It serves no purpose for everyone else. Meaning its very hard to convience anyone to fix it without someone else paying.

The only thing silly is to believe that companies like EA etc would happy throw money after a patch so a single new GPU uarch could run better with Mantle. Can you hear the laughter? Yes, thats the EA executives you hear. Same applies for any other company. They are not going to deliver "welfare" patches for AMD or any other company for their blue eyes sake.

 

[tviceman]

Good review. Thanks Ryan Smith! Your analysis is always top notch. I hope your punctuality (in the future) returns to form!!!