D700s Today...

[koyoot]

The D700 is a 7970 that has had its core clock reduced by 8% to fit the thermal profile of the nMP. The number of shader processors, ROPs, and TMUs are exactly the same as the reference desktop 7970 board. The memory clock and memory bus width are the same. The D700/7970 is well out of date but in no way is it a mobile part.

To me, these bizarre Nvidia-fanboy rants of yours over the last few weeks are really undermining your credibility.

Also its worth noting that Mobile parts are usually best binned parts. If anyone would want to know what that means, here is full article.
https://en.wikipedia.org/wiki/Product_binning
D300 is 7870, or R9 270X chip(Pitcairn) in a 136W TDP power envelope.
D500 is cut down Tahiti(7970) chip. In everything, it means that Mobile parts are in some way better than desktop parts, because they can much better maintain performance in lower power-thermal envelopes.

 

[Fl0r!an]

That would be because the reference 7770 was designed to be a low-ish power chip in the first place, probably with an eye on top-end laptops. Even at the full 800MHz clock it only uses 75W and takes all the power it needs from the PCIE slot.

Same applies to the R9 M290X in the 5K iMac: Full-size, downclocked Pitcairn chip.

I don't really see what this discussion is about. Obviously you will never see any notebook carrying 2 D700s as it would need nitrogen cooling and it's own nuclear power plant, so the term "mobile GPU" is a little misleading. The process in creating those GPUs (Dx00s vs. mobile) is still similar though.

 

[netkas]

We can't stop the debates easily

Here is interesting results, R9 390x @ 1100 Mhz more or less equals to GeForce 980 in performance at high res (4k and 2560p ) but consumes 137 watts of energy more. - http://hexus.net/tech/reviews/graphics/85436-powercolor-radeon-r9-390x-devil/?page=11

Something is very wrong with amd 28nm chips at clocks of 1ghz and higher.

The chips in macpro and imacs runs around 800 mhz for a good reason, they are able to run at low volts at 800mhz those consuming much less power.

 

[koyoot]

Netkas, not exactly.
http://tpucdn.com/reviews/Powercolor/R9_390_PCS_Plus/images/crysis3_1920_1080.gif
http://tpucdn.com/reviews/Powercolor/R9_390_PCS_Plus/images/bioshock_1920_1080.gif
http://tpucdn.com/reviews/Powercolor/R9_390_PCS_Plus/images/power_average.gif
http://tpucdn.com/reviews/Powercolor/R9_390_PCS_Plus/images/power_peak.gif
R9 390 is already at stock clocks faster in some situation than GTX 980 even in 1080p resolution.

It really depends on situation and on the game or task. Its also worth noting that R9 390 at stock clocks has lower power consumption, than R9 290, even if it has higher clock on core, higher clock on memory and twice the amount of VRAM. Only if you push the GPU much over its nominal power envelope you get much higher power consumption.

It is the exact difference between Asus Strix Fury and Sapphire Fury. Strix has 1000 MHz core clock whereas Saaphire has 1040 MHz on core with the same voltage as Fury X. What is the result is that Asus is within its BIOS cap of 216W and Sapphire goes all the way to 330W in power consumption. http://www.anandtech.com/show/9421/the-amd-radeon-r9-fury-review-feat-sapphire-asus/17
Here is Anandtech complete analysis which backs up your words in Mac Pro 2015 thread, and also mine words about downclocking and downvolting.

The end result is that the Sapphire card has a much wider range than the ASUS card, and it also indicates that the ASUS card’s power limit is likely not very far above what it takes to sustain 1000MHz under most games. The ASUS card will likely need more power to avoid power throttling if overclocked.

Exactly the same thing we see in D700 and Mac Pro. It is not able to OC, because of the voltage and power target.
So at -43 mV on core we get 40 MHz less on core clock, fully stable and maintainable in normal circumstances to achieve full potential of a GPU. 3584 GCN cores x 2/core clock = TFLOPs performance. 7.168 TFLOPs of compute power from 216W of power. I would like to point at Nvidia offerings here. GTX 980 Ti - 5.6 TFLOPs. Titan X - 6 TFLOPs at 250W TDP power target.

http://scr3.golem.de/screenshots/15...uxmark-v3-(complex-scene,-gpu-only)-chart.png And yet, still slower and weaker than R9 290X.
http://scr3.golem.de/screenshots/15...uxmark-v3,-complex-scene,-gpu-only)-chart.png
While using similar amount of power. How much faster and more powerful would be Fury while using much less power than Nvidia Offerings? .
http://s28.postimg.org/a8bdgk7pp/Capture.png
http://s30.postimg.org/431cr5ej5/Capture.png
http://s28.postimg.org/4vr4g6971/Capture.png

 

[netkas]

you are completely wrong about asus r9 fury, it consumes around 260 w under load, way more than 216w limit they set in bios.
image from your link, asus uses just 30 watts less then sapphire.

http://images.anandtech.com/graphs/graph9421/75693.png

learn to read what you post please.

 

[koyoot]

Asus Strix Fury.
http://tpucdn.com/reviews/ASUS/R9_Fury_Strix/images/power_average.gif
http://tpucdn.com/reviews/ASUS/R9_Fury_Strix/images/power_peak.gif
Sapphire Fury
http://tpucdn.com/reviews/Sapphire/R9_Fury_Tri-X_OC/images/power_average.gif
http://tpucdn.com/reviews/Sapphire/R9_Fury_Tri-X_OC/images/power_peak.gif

Both sites has the similar test for power consumption. TPU used Crysis 3 for very long time, now they are using Metro: Last Light.

http://scr3.golem.de/screenshots/15...uxmark-v3,-complex-scene,-gpu-only)-chart.png

If Fury X with very high voltage and clock consumes 249W while OpenCL work, how much less power will Asus Strix Fury consume with voltage, power, and clock cap?

 

[netkas]

It's questionable how tpu got that numbers:

For this test, we measure the power consumption of only the graphics card via the PCI-Express power connector(s) and PCI-Express bus slot. A Keithley Integra 2700 digital multimeter with 6.5-digit resolution is used for all measurements. Again, the values here only reflect the card's power consumption as measured at its DC inputs, not that of the whole system.


I would believe measurements at wall plug done by anandtech more.


Opencl tests dont' utilize all units in gpu, tmu/rops aren't used much.

For example geForce titan would pass opencl test on internal-only power and would shutdown instantly in furmark and in some heavy 3dmark firestrike ultra tests.

 

[koyoot]

It is still consistent with what you have posted. 30W difference between both cards in both scenarios. Exactly in line with the power consumption on Anandtech site.

And no, It is not questionable.