AMD "Fiji" HBM Implementation Detailed

[RejZoR]

I'll be getting new generation out of curiosity. i frankly don't have any real need with overclocked HD7950 3GB. Anything I throw at it works like a charm. Except maybe Natural Selection 2 which is very demanding game on CPU and GPU, especially in late games when there are tons of buildings constructed. I just hope they'll shift to DirectX 12 so we'll offload the burden from CPU. I mostly buy new cards for cool features more than for raw performance. For example I wanted HD6950 over HD6870 because the 6900 supported EQAA and HD6800 didn't. Next one was MLAA and R9-300 will most likely be because of HBM, VSR and other goodies.

 

[2wicked]

First thing I thought of when I read about the way they're packaging the chip + mem.

 

[Casecutter]

This is what I've been always saying, especially since I work with customers on daily basis and have seen it first hand during a struggle in a company where I work few years ago. We used to have like 15 different models for the same product type and it was really difficult to sell them, people often had to go and "rethink" despite literally pushing them towards one product. But then we were forced to narrow down the lineup to like 5-6 models, basically 2 for each price range. Guess what, sales went through the roof almost, because people were able to quickly decide what is within their budget and what isn't. And when they had to pay more, it was also an easy choice. The first more expensive one. But if you granulate the models to 10 extra models in between, you confuse them again with decisions how much extra is worth paying for what extra features.

And that's what baffles me with AMD's Radeon product lines. There are TOO MANY of them. Bunch of Rx editions and then those are granulated down to bunch of series and then down to special models and versions. Totally unnecessary, confusing and overwhelming for costumers.

Why not have just R9 and place 4-5 well placed models here. Simulated naming for the new R9-300 series:

R9-320 2GB (budget)
R9-350 2GB (low end)
R9-370 3GB (mid range)
R9-380 4GB (high end)
R9-390 4GB/8GB* (premium-enthusiast)

*depending on how HBM can be implemented currently

No X versions, no LE crap versions, no various memory configurations, no R5 and R7 unless if you want to separate lets say mobile chips from desktops that way. When you give people a good argument why they should pay few bucks more, they will do so. But if you're having hard time justifying why every iteration of a card in between costs X for function Y, you just confuse costumers and make them walk away because they have to "rethink". And everyone who has to rethink is more likely to buy something from competition.

I don't know, I don't have a science degree in space marketing and a 6 figure yearly income and I get this. But companies just go with heads straight through the walls. Go figure...

Whole heartily agree on what you say, the problem is manufacturing process will almost always furnishes you less than perfect chips, that you need to use-up to achieve any viable pricing. Although, the maturity of 28nm has been good with geldings binned more by clock less by non-functional areas on the die (as say in the case of the 270/270X). There’s still huge spans in the segments that calculating marketing types hate not have a footing in.

As to the budget/low-end yes AMD has way too much confusion in that range! They need just a budget/basic that's more a repair or diagnosis type card even if it a from two process’ ago (6450). Then as "low-end"; a nice step above a common iGPU in mainstream CPU, good HTPC and low power. Ultimately with AMD that “Low-End” should be intended pair up with APU's to really provide "Dual Graphics" (for once) that offers enhanced "entry gaming" (gets you into 1080p) when paired together. That really needs to be a compelling straight-forward upgrade, for APU OEM boxes. AMD should be off-loading APU’s in the market for margins even if razor thin, and really work to pick up that discrete card sale of $60 after.

After that there's:
Entry: Adolescence-Early teens, $80-120 works with OEM 300W PSU
Mainstream: Teen's (6-10 hrs/wk) strong 1080p, $130-200
Gamer: (12-25 hrs/wk) High settings 1080p, able to provide >med. 1440p, ≥500W PSU $230-500
Enthusiast: Whatever...

So you really end up designing a chip for each of those 4 segments, and being there's the full-fledged and a least one gelding, quickly you could see at minimum 6 cards, if you get lucky and the cost/perf/power works to transfer to the lower level. It's nice if your gelding parts can become mobile/OEM discrete, but then such market don't gobble the volume of chips that you amass if doing really great on the retail side.

It’s more of a juggling act between what engineering finally provides, where the competition falls perf/$, how long can you bin/hold geldings and will there be a good opening to exploit if you sit on them. It’s tougher than folk think, especially if you’re the underdog and every dollar counts.

 

[Aquinus]

I'm feeling a short dual-GPU solution in AMD's future. In fact I'm seeing a lot of very short video cards if memory is removed from the graphics card PCB. It's kind of like how SoC's allow for really small motherboards.

I was thinking about replacing my 6870s, but I think I'm going to hold on for a little bit for obvious reasons.

 

[PLAfiller]

This is what I've been always saying, especially since I work with customers on daily basis and have seen it first hand during a struggle in a company where I work few years ago. We used to have like 15 different models for the same product type and it was really difficult to sell them, people often had to go and "rethink" despite literally pushing them towards one product. But then we were forced to narrow down the lineup to like 5-6 models, basically 2 for each price range. Guess what, sales went through the roof almost, because people were able to quickly decide what is within their budget and what isn't. And when they had to pay more, it was also an easy choice. The first more expensive one. But if you granulate the models to 10 extra models in between, you confuse them again with decisions how much extra is worth paying for what extra features.

And that's what baffles me with AMD's Radeon product lines. There are TOO MANY of them. Bunch of Rx editions and then those are granulated down to bunch of series and then down to special models and versions. Totally unnecessary, confusing and overwhelming for costumers.

Why not have just R9 and place 4-5 well placed models here. Simulated naming for the new R9-300 series:

R9-320 2GB (budget)
R9-350 2GB (low end)
R9-370 3GB (mid range)
R9-380 4GB (high end)
R9-390 4GB/8GB* (premium-enthusiast)

*depending on how HBM can be implemented currently

No X versions, no LE crap versions, no various memory configurations, no R5 and R7 unless if you want to separate lets say mobile chips from desktops that way. When you give people a good argument why they should pay few bucks more, they will do so. But if you're having hard time justifying why every iteration of a card in between costs X for function Y, you just confuse costumers and make them walk away because they have to "rethink". And everyone who has to rethink is more likely to buy something from competition.

I don't know, I don't have a science degree in space marketing and a 6 figure yearly income and I get this. But companies just go with heads straight through the walls. Go figure...

I couldn't have said it better. Thanks for sharing some experience.

Really nice featurette in general and pretty notable comments on the topic. Despite being in the "green camp" almost all my life so far, I must say I am intrigued about what AMD is cooking. I don't know if I will jump on the wagon, but I'd like them to have something low-profile-sized with DDR5 for around a 100 bucks that's actually better than GT740/640 for my portable RIG.