I’d love to print my own enclosure, I just don’t really know how to work in CAD, and I’m still saving up for my own printer. My plan was to order a print of the provided VESA mountable tray and somehow mount it to the side of a drive cage like this one from Rosewill. But I’d love to design a more elegant solution.
I’d been looking for some sort of USB 3.2 bay but you’re spot on with the SATA adapter - I had no clue they existed. I guess then I’d just need to sort out power delivery and then figure out the enclosure.
ah yeah. CAD can be quite daunting.
You don’t have to get a printer. You can get your parts printed.
Especially if it’s just a few small prints. If you 3D print like 3 piece a week (for a total of 30 pieces or more) I think a printer will be worth it.
My Institute use Siemens NX so that’s what I uses. I tried FreeCAD and it’s just nightmare. I think it will be fine for someone fresh to pick up, but I frankly can’t even find where I will make a 2D drawing out from.
Would anyone else with a 1135g7 mind posting their cinebench23 results? With all my apps closed, windows update disabled, and bios 3.10 installed, I was only able to achieve a score of 5003.
Not sure what makes the difference here, but the highest single-run Cinebench R23 score I got with the 1135G7 was: 1296 single, 5408 multi
And now that I’ve upgraded to the 1240P…
I got 10115 for single-run and 9467 for 10-minute on my 1240P. Since their score is 9631… it could go either way?
EDIT: Just read their post more, it says “A Cinebench run brings the CPU to maximum load for 10 minutes, making it a great measure of sustained performance.”
My bet is, this is caused by DDR4 memory instead of faster LPDDR4X / LPDDR5. GPUs require very high memory bandwidth to be fully utilized.
And while I value the modularity and upgradability and so far am happy with my 12th gen FrameWork, it would be great to have some apples-to-apples comparisons to confirm my theory or just put numbers to certain design choices, like what the cost in terms of efficiency and performance is for that modularity and allowing me to reuse my existing DDR4 So-Dimms.
And if I am wrong and it makes no appreciable difference in practice then even better.
I would have found it a little bit fairer if the tests where done on machines with the same RAM capacity. I am no export by any means and this shines now through. How can I know that the performance gain is actually due to CPU and not RAM as well?
Maybe these games do not need that much RAM though…
Correct me if I am wrong, but I do not believe the iGPU was used in those benchmarks.
I understand that the choice between DDR4 and DDR5 is negligible in most applications. But iGPUs have been historically very memory bandwidth sensitive and benefited greatly from dual channel and higher memory clocks. But its very hard to come by numbers. Especially when different notebooks are compared, it is very hard to remove the impact of cooling.
Make sure you are plugged in to a power supply than can do 60W (it might not turbo without) and make sure the power profile is set to Performance (not Balanced).
I’d be willing to bet that there will be a new one around the time that third-party reviews come out and shipping starts, since that’s when this one was posted.
