3 ms·

> I don’t understand how my desktop i5 4690k with an old SATA SSD is SO much faster than this laptop that has an 8th gen i7 8650u, DDR4, NVMe SSD, even for basi

> I don’t understand how my desktop i5 4690k with an old SATA SSD is SO much faster than this laptop that has an 8th gen i7 8650u, DDR4, NVMe SSD, even for basic tasks like having more than 20 browser tabs open and just surfing the web. The surface book is in a strange bind. the 8650 is much much weaker than your 4690k and has very little thermal room to play with. The thermals are particularly tricky because the CPU lives in the tablet portion; meaning there is a tension between cooling it properly, providing a touch screen and keeping it thin and light enough to not have awkward weight distribution.

madjam002 6y ago
Yup that's what I've found, unfortunately for me I realised this after purchasing it. It's odd that it's marketed as a high end performance laptop whereas it's more comparable to a low-mid range desktop.
- dr_zoidberg 6y ago
  Just from the power envelope (88W for the desktop vs 15 for the 8650u in the surface book 2, assuming it's configured to that setting -- edit: it is) the desktop part would be 80% faster, if all things were equal. Now all things aren't equal. The mobile chip is going to be thermally limited, while a desktop chip has a huge heatsink and fan to it's service, and a lot of room to be "just warm". Despite whatever architecture improvements (and take those with a grain of salt, remember that skylake and its descendants are the most affected by the Spectre/Meltdown mitigations) you still have physics against you and the desktop part should be 50% faster simply because it can keep the turbos up and higher frequencies than a TDP-restrained mobile processor. It'd seem that, today, only the AMD Renoirs are capable of pulling "desktop-like performance on a mobile budget". Let's hope AMD keeps bringing good chips out, and Intel wakes up and stands with something decent, unlike what they've been doing so far. -- edit: Since I was already editing: of course a Ryzen whatever is going to be also TDP-limited compared against other Ryzens, but AMD isn't really making the mobile parts completely equivalent to the desktop parts. So yes, calling a Renoir "desktop performance" is massively ignoring AMD "super powerful" desktop chips. More reason to hope Intel gets up and starts giving a fight here. We need competition, otherwise AMD will turn the tables and will get back to a stagnant situation.
- ric2b 6y ago
  > It's odd that it's marketed as a high end performance laptop whereas it's more comparable to a low-mid range desktop. That's true of nearly all laptops except maybe bulky gaming laptops. The thermal and power limits of laptops mean that they're not really "high performance", they're just high performance for a laptop.

madjam002 6y ago

Yup that's what I've found, unfortunately for me I realised this after purchasing it. It's odd that it's marketed as a high end performance laptop whereas it's more comparable to a low-mid range desktop.

dr_zoidberg 6y ago
Just from the power envelope (88W for the desktop vs 15 for the 8650u in the surface book 2, assuming it's configured to that setting -- edit: it is) the desktop part would be 80% faster, if all things were equal. Now all things aren't equal. The mobile chip is going to be thermally limited, while a desktop chip has a huge heatsink and fan to it's service, and a lot of room to be "just warm". Despite whatever architecture improvements (and take those with a grain of salt, remember that skylake and its descendants are the most affected by the Spectre/Meltdown mitigations) you still have physics against you and the desktop part should be 50% faster simply because it can keep the turbos up and higher frequencies than a TDP-restrained mobile processor. It'd seem that, today, only the AMD Renoirs are capable of pulling "desktop-like performance on a mobile budget". Let's hope AMD keeps bringing good chips out, and Intel wakes up and stands with something decent, unlike what they've been doing so far. -- edit: Since I was already editing: of course a Ryzen whatever is going to be also TDP-limited compared against other Ryzens, but AMD isn't really making the mobile parts completely equivalent to the desktop parts. So yes, calling a Renoir "desktop performance" is massively ignoring AMD "super powerful" desktop chips. More reason to hope Intel gets up and starts giving a fight here. We need competition, otherwise AMD will turn the tables and will get back to a stagnant situation.
ric2b 6y ago
> It's odd that it's marketed as a high end performance laptop whereas it's more comparable to a low-mid range desktop. That's true of nearly all laptops except maybe bulky gaming laptops. The thermal and power limits of laptops mean that they're not really "high performance", they're just high performance for a laptop.