Using BIOS 1.27 in my p1g2 with the slower i7-9750H, I got 1010 in Cinebench R15 multi core. After undervolting the CPU by 130 mv (a five minute software tweak that doesn't void the warranty) I averaged 1243. Aain that's with the slower 9750H CPU.

It would be nice to have information about the firmware/BIOS version, and see a new test if any upgrade was possible.

Have you updated the firmware and re-evaluated? It seems the throttling was overly aggressive which might explain why several commenters aren't seeing the same.

I have to say, this post worried me a bit. I read it for the first time a day after ordering my own P1 Gen2.

I just received it, and I'm not getting results like you've posted. My average on CB R15 seems to sit around 1045, with a high of 1086 and a low of 1015.

My setup is:
i7-9850H 6c 12threads @ 2.60GHZ
Quadro T2000
16gb ram at 2666

@IO,
no. The memory makes no difference for the Cinebench test and the configurations are extremely comparable. The Core i7-8850H is by no means a "basic" configuration.

You are comparing an older high end configuration with a Xeon processor and faster memory to a basic configuration. This is not very informative, try comparing similar configurations...

Quote from: ormandj on August 22, 2019, 14:35:19

QuoteThus, we were disappoint to find out during our review that the ThinkPad P1 2019 actually lacked behind its predecessor, the ThinkPad P1 Gen 1.

QuoteStill, as visible in the graph above, it still lacks behind the older ThinkPad P1 2018.

"lack" should be replaced by "lag" throughout the article. s/lacked/lagged/ and s/lacks/lags/

Interesting find, it's frustrating as a consumer to not be able to rely on specifications of these laptops due to the manufacturers inability to keep them cool. It would be nice if we could settle on some standardized protocol for testing performance that manufacturers would be required to run and issue with their laptops. Just like EPA mpg ratings, it wouldn't be perfect, but it'd be better than it is now.

We appreciate the early results potentially helping avoid spending on devices that don't fill needs due to poor thermal control.

Intel should probably be apportioned most of the blame, not laptop manufacturers. Due to Intels dominant/monopoly position on laptop CPU's plus the fact that are still stuck on 14nm transistor size, their CPU's have had their thermals skyrocket (Intel should have been on 7nm by now but got delayed by 5 years).

Intel then tried to "offload" this problem onto laptop manufacturers by giving them a lot of dials and nobs to try and intelligently tune their CPU but these are just workarounds (and in many cases to get decent thermals out of new Intel CPU's you just end up underclocking it to be as fast as previous generations).

I mean this makes sense, if you cant reduce the size of the transistors then the only way to significantly improve performance of a CPU is by adding more transistors which takes up more space/power and generates more heat. If only we had AMD 7nm CPU for laptop line......

how about vantage settings and the CPU's thermal cap? read more about it here: https__www.reddit.com/r/thinkpad/comments/cpiwcy/x1extreme_cpu_thermal_cap_at_80c/

QuoteThus, we were disappoint to find out during our review that the ThinkPad P1 2019 actually lacked behind its predecessor, the ThinkPad P1 Gen 1.

QuoteStill, as visible in the graph above, it still lacks behind the older ThinkPad P1 2018.

"lack" should be replaced by "lag" throughout the article. s/lacked/lagged/ and s/lacks/lags/

Interesting find, it's frustrating as a consumer to not be able to rely on specifications of these laptops due to the manufacturers inability to keep them cool. It would be nice if we could settle on some standardized protocol for testing performance that manufacturers would be required to run and issue with their laptops. Just like EPA mpg ratings, it wouldn't be perfect, but it'd be better than it is now.

We appreciate the early results potentially helping avoid spending on devices that don't fill needs due to poor thermal control.

Could the fact that the 1st Gen P1 used has a Xeon processor play a role?  I thought the Xeons were supposed to have better stability for sustained multi-core workloads.

Regardless, it does seem reasonable to question whether 9th-gen CPUs really offer much benefit over 8th-gen.  Hopefully Ice Lake will see a nice spike.

Hi, I do not have the same results at all.

I have benchmarked a P1 gen 2 (9850H, T2000) extensively against my x1 extreme gen 1 and in every benchmark I have run the P1 gen 2 comes out ahead. These include cinebench r15 (P1 averaged > 1000 over 3 runs with x1e never scoring higher than 750), 3dmark timespy (CPU and graphics scores are higher) and geekbench also significantly higher. Is there variance? Yes, apparently. But your title does not at all match with my experience...