Precision 7530 & Precision 7730 owner's thread

That's because Dell is lying.

Same thing with P3200 in 7530. No Max-Q branding despite being even lower TDP/slower than P3200 Max-Q in Lenovo P52.

 

Man, i respect you but it doesn’t matter how long you are here. What really matters is the fact. You use my screenshots to defend your own (actually incorrect) opinion, this isn’t a great idea.
1. Please post your own screenshots to prove to us all p5200 can maintain above 125w. Show the power and show the 3d mark score please. If your p5200 is a full power version then we will be more than happy to know it, and we can ask dell for an exchange.
2. Why by looking at my screenshot you got the conclusion that it can maintain 125-131w all day? I am really confused.
3. Actually in 51nb forum, there is another guy with 8950+p5200, his results are almost the same as mine. We can both confirm, at least for now, p5200 on 7730 is around 110w. According to notebookcheck, 110w p5200 is usually what we call a max-q version.

Please check notebookcheck's description, here I post it again:

"
For the Max-P variant, the clock speeds are specified from 1556 (base) to 1746 MHz (Boost), the power efficient Max-Q clocks from 1316 MHz (base) to 1569 (Boost) according to a GPU-Z screenshot from a reader (-15% / -10%). The 16 GB GDDR5 are clocked at 3,6 GHz leading to a memory bandwidth of 230 GB/s.
"


As you can see from my photos, the core is about 1404Mhz, I have never seen it above 1500Mhz at all, not to say 1700mhz. It is just around 1400mhz. And, considering its power is around 110w. We have reason to believe this P5200 is indeed a max-q verison. not the full power 150w version.

 

You are right man.

 

Bokeh did not prove anything. He used my screenshots in another forum to teach me that my p5200 can maintain above 125w. I really feel very funny about this.

 

I'm suprised by the attitude towards Bokeh. I would have expected it to be in the spirit of "that's interesting, let's find out why you have a different result than we do" rather than "that's false, give us proof".

Also, CR3, to respond to your earlier remark regarding power usage in earlier generations: it does not make sense to expect modern day devices to consume more power than older models. If anything, the opposite should be expected. After all, the goal is to gain as much performance per watt as possible, not to build a notebook that requires a private nuclear reactor. Complaining that an earlier model could spend 68W on a CPU and the current gen only spends 60W is like complaining that modern cars don't have the same excessive fuel consumption as the gas guzzling models from the 50's.

 

I didn't see any proof from Bokeh or indication that he owns the system in question. Using data gathered from a different machine using a GPU with a different power limit is not proof.

The discussion about power limits is very relevant because this generation added more cores. A six cylinder should have higher fuel consumption than a four cylinder.

 

But all information when have in any forums shows the GPU is around 110 W. and CR3 is the one shows a solid proof

Ask for proof is actually the right way to end meaningless argue.

 

The issue is not what Watts are "spent". It is what they are limited to artificially via either firmware, or implementation of Nvidia's Max-Q line (purposely weakened GPUs).

The current gen of Pascal Quadro GPUs already have known TDP (Watt consumption/thermal design) and are based off very specific core architecture, core count, memory bandwidth, and GPU frequency. All this information is released on announcement of the GPU.

If Dell is either artificially limiting the TDP and GPU frequency themselves, or using the MAX-Q (Must be branded as such based off NVIDIA rules); people deserve to know.

To use your car analogy:
You like to buy a certain sports car. It comes in 2 versions, one with the normal advertised V8, another with lower power V6 for fuel efficiency.
But the dealership will not tell you which version is in the car you are looking at; and they charge the same high price for both. Wouldn't you be irritated?

 

Friend, i want to make things quick and direct. Like, i don’t like spending time saying things like, how are you, how do you do, how’s day been... i want to be direct, go to the point. If i was wrong i surely apologize. i said i did not mean to offend anyone. I want a fact, simple and pure.

About power, sorry i do not agree with what you said. We not talking about cellphone. We are talking about precision 7530/7730 which targets performance. Sure, of course everyone knows that we need more performance per watt, everyone knows that. I am not talking about this at all.
What i want to say is, if in 2013 we can give 15 inches m4800 cpu 68w and the system can handle it, then why, in 2018, for the flagship 17 inch 7730, we only give it 60w? Of course the 60w today gives more performance than a 2013’s 60w cpu, everyone knows that. But why not giving it 65-70w? Why so mean at 60w? 70w or 60w, it doesn’t make much differece for electricity bill per month, but the performance of 70w will be convincingly better than 60w. So why not try 70w instead of 60w?
I mean it is like, i buy a truck of course i know it eats more fuel. But i need a truck because of its power. Otherwise i would just buy a small Toyota instead.

 

Where did I post on another forum? Was I replying to your post and it copied your screenshot? I have posted a GPU-Z screenshot that I took on another thread here at NBR, but that was a pic I took in Las Vegas as a Dell Technologies World attendee.

But, I did mess up when I said I had posted a screenshot showing the 125w+ on the P5200. I went back and I had only posted the HWinfo64 screenshot of the i9 running at 5.2ghz. I thought I had posted both of them, but didn’t.

My Dell rep set it so that I was able to look at two 7730 systems. System 1 had the i9 with a P3200 and system 2 had the xeon with a P5200. I was able to run tests on both of them before returning them, but had access to the i9 system for much less time, so I just focused on how well the i9 overclocked compared to the AW17. With the xeon/P5200 system, I focused on the gpu performance relative to the 1080 OC in the AW17 and an MSI 1070 desktop gpu.

I will try to get my hands on the P5200 system again.