I am gonna need some popcorn here....:p:p

Fox2232:

We were already there, weren't we? You just skipped cost analysis of the chip as even rough approximate would make Navi quite cheaper.

I'm honestly not even following you on this one

Like Turing again? So many more transistors for compute, little gaming improvement. Compute is not free. Just because there is chip that has even more transistors does not mean that 2080Ti can be that strong without that transistor investment.

Now you're just being disingenuous. the added hardware of Turing is obviously very specialized. And V100 is a perfect example of how adding compute hardware isn't an excuse for why AMD's GPU's are so inefficient. That hardware is there, and it's till a more efficient gaming GPU than Vega 20 is, while being on a larger node.

So, they did 10% efficiency improvement year and half ago, and then called it a day waiting for 7nm, right? So we are in incompetent AMD category. Spinning class 101.

AMD's GPU division has been in incompetent territory for years now.

Edit: And just a side note: I would love to be in AMD's room when they decided on: "This is going to be last GCN iteration. Lets spent a lot of time and resources to make as good as GCN can be before we drop it for no further benefit." Then 2 and half years later another meeting room: "Well, we made this last GCN hurray, what about spending another year on it to fine tune everything?" I would be laughing at that moment right into their face. Because such investment into something that has no future should make everyone in that room laugh too.

right, because missing goals and unpredictable fab schedules aren't a thing. We'll all see soon enough. I'll enjoy reminding you how this turns out.

Fox2232:

V100 has 59.5% more transistors than Radeon 7 to have 5% higher FP16/32 and just double FP64. Want to take Pro cards here? Then lets take in account that Radeon 7 is Cut down and disabled MI60. AMD's MI60 is actually using all of those 13.23 Billion transistors And is faster in all FP16/32/64 than V100 that has 21.1 Billion transistors.

Yeah but V100 has tensors that are dedicated solely for INT8/INT4 and it more than doubles the performance of both operations compared to the MI60. MI60 has 59TOPS of INT8 and 118TOPS INT4 - V100 has 130 and 260 respectively. Now I doubt those make up the entire transistor difference but it's probably a fair percentage of it. That being said I think AMD's GPU division is incredible competent given the budget/financial restrictions they are operating under. GCN has scaled extremely well and their efforts to "steer" the industry with consoles towards their benefit was a pretty good call on their part. I'm excited to see what Navi brings honestly.

Fox2232:

Sure, and that's the cost to be paid in transistors which ttnuagmada refused to understand.

You continue to miss the point. My point this entire time is that AMD can't add more transistors. It's design is too inefficient. That is AMD's entire problem. That is why AMD is so far behind.

Fox2232:

That's baseless. AMD can't really add more transistors to GCN because only way to do it is to make CUs beefier. (As they can't put more than 64CUs there due to design limits in logic.) If GCN could have 96CUs, they would just decrease clock by 10%, reduce voltage appropriately to lower max clock and be at same power draw as now. And it would still be smaller GPU than nVidia's counterpart that it would compete with while having much better compute than it already has. (Here you can say that AMD made unbelievably shortsighted choice. Their 1st GCN chip already had 32CUs, that's 1/2 of maximum.)

I'm afraid it doesn't work that way, as a quick glance at any perf/watt chart from a techpowerup review will clearly show you. AMD would need 400w at a minimum to make exactly what you described.

Fox2232:

Same way as any other AMD's GPU being sold at unreasonably clock and voltage. According to TPU database Fury nano is just 10% slower (1080p) than Fury X while Nano has 175W TDP and tiny cooler and Fury X has 300W TDP and integrated liquid cooling. Same way as I was able to reduce power draw of any AMD's GCN card by good 25~30% with minimal performance loss or gain if I tuned memory timings and clock. Sacrificing clock and reduced voltage will get you very far. Have you ever seen AMD's Zen based APUs with Vega iGPU? Looks like you did not. Kind of funny if you look at their performance at 15/35/65/95W. There's really not that big of an difference.

Ahh, of course, the ol "undervolting is an AMD only feature" argument. ignoring the fact that literally all semiconductors behave the same way, and that the reason they all mostly run at higher volts than they need, is because there are always a few that can't. AMD owners are just obsessed with it due to their dumpster tier efficiency at stock settings. Both of my 1080ti's will do 1800mhz at 800mv. Undervolting is not AMD specific.

ttnuagmada:

Ahh, of course, the ol "undervolting is an AMD only feature" argument. ignoring the fact that literally all semiconductors behave the same way, and that the reason they all mostly run at higher volts than they need, is because there are always a few that can't. AMD owners are just obsessed with it due to their dumpster tier efficiency at stock settings. Both of my 1080ti's will do 1800mhz at 800mv. Undervolting is not AMD specific.

Although to be fair, compared to how much AMD cards ship with a higher voltages than needed compared to Nvidia is quite high to the point where if you have an AMD gpu you almost have to tune the voltage for it to be reasonable. Nvidia cards out the box IMO at least have reasonable voltages where you do not need to tune them at all. Not sure how undervolting is on Turing, but at least with cards like the Radeon VII can be undervolted to be competitive in power draw compared to the competition.