Hands on Dell Precision 7710

I think their point was it's helpful to be able to switch back to the original drive when troubleshooting OTHER components with Dell, to verify a hardware failure later on.


Sent from my iPhone using Tapatalk

 

You should definitely keep the original drive around, Dell has the option to require that you switch the hardware back to the original configuration when troubleshooting warranty issues before scheduling service.

I've never actually experienced this and I typically do swap some parts in my systems, but I've heard specifically from Dell people that this is an option for them, whether or not they require it depends on the particular issue and on the tech handling it.

 

Hi guys, this is the configuration I'm at, looking good? https://gyazo.com/a5f1de99e6cce51eaa4abecce62ce144
Going to talk with live chat and work on getting a discount. I think I'm also going to get the Timbuk2 Premium Messenger Case and the Premium Lock LP500.

 

I have been able to dig up a benchmark for the forthcoming Xeon E3-1545m:

http://www.passmark.com/baselines/V8/display.php?id=57471132163

Assuming it's legit, the tested E3-1545m seems to be about 7% faster than than the average E3-1535m. 3D Scores seem to be a bit on the low side for an M5000M, but that might be a power setting or driver issue.

 

the E3 mobile xeons are also listed on the Intel Ark

http://ark.intel.com/compare/89608,89610,93359,93354

the main difference between a 1535 and upcoming 1545, according to Intel, seems to be the Integrated Graphics. Note the improved resolution using WiDi on the 1545 compared to the 1535 and 1505. I didn't even know about WiDi before I did compare.

Other than integrated graphics, not much different. "7% faster" could be from many different variables and I am not sure if that is absolute.

 

The interesting thing is that 1535m and 1545m have the same clockspeeds (base 2.9 GHz, Turbo 3.8 GHz), yet the 1545m still seems to come out on top.

That's probably due the 1545m having twice as much eDRAM (128MB) as the 1535m (64MB). Skylake uses the eDRAM as "don't-call-me-L4-Cache" DRAM buffer.

Anandtech has a nice explanation how the eDRAM in Skylake works: http://www.anandtech.com/show/9990/...tels-pricing-lists-xeon-e3-1575m-v5-with-gt4e

Since the 1545m is just 9% more expensive than the 1535m according to official price list, it might be worth the wait.

BTW Intel is also offering non-Xeon mobile Skylakes with GT4e graphics. The top model is the i7-6970HQ, which is a tiny bit slower than the 1545m @2.8/3.7 GHz: http://ark.intel.com/products/93336/Intel-Core-i7-6970HQ-Processor-8M-Cache-up-to-3_70-GHz

 

that is interesting. Sounds like Iris Pro is more than just an IGP upgrade. I'm really curious about the 1575 and why it is double the price.

 

PCmark CPU test is also influenced by the hard drive. More likely, the 1545 machine had a faster solid state drive thus giving a 7% increase, which isn't much considering the general rule of thumb is that you won't notice anything under 10% increase in day to day work.

My other concern is where did they get the 1545M? There's no machine currently available with it so the results could have been faked as well.

 

Well, these were not PCmark figures but Passmark PerformanceTest figures (http://www.passmark.com/products/pt.htm) and judging from the description and from the fact that there is a separate HDD test in the Passmark software, I doubt that the HDD influences the CPU scores.

I'm not familiar with the Passmark software, so I don't know if it is possible to submit faked results. But according to the Passmark website the test was run on February the 21st 2016. Since the 1545m and the 1575m appeared on the Intel pricelist at the end of january, there could be pre-release machines now. The timeline fits. Someone could have run the test on a pre-release machine, maybe as a stress test or as part of the certification

 

Whoops, I meant passmark. Look at the different CPU scores for the same 7710s with 1535M. They vary greatly based on the solid state drive being used