Laptop CPUs typically target around 1W idle. 5W is more typical for the entire system. That results in around 11h battery life.

I used to have asus turbo gtx970. It was a bit loud but worked great in a small case.

I now have the gigabyte eagle 3060ti and I accidentally had power limit set to 150W and didn’t even notice anything until I started to look at why it’s so quiet in heavy games. Blower would be very nice for that level of power.

Motherboards cost a lot mainly due to cost of all the layers needed for routing and integrity of high speed signals. Also power management.

He’s Chinese so that’s just a normal workday.

Nvidia stock has been priced for about tenfold growth already. Unless they literally grow 1000% there is little reason for the price to move.

That would require on chip integrated voltage regulators per core so no, they only do that on servers. There will be voltage regulators per tile and bus at least though.

That’s not really a question but a complaint. Your reformulation of it as a question boils down to “will next gen perform better than the previous gen” to which the answer is yes, it will.

In 99% of use cases you get exactly what benchmarks promise. Power limits only affect very small set of workloads.

Not really in most cases. The decoder might need to spend some more transistors to accommodate the instructions but that should not be much. And the very oldest never used ones can be thrown to some very slow microcode rom or something. In the execution side SSE uses the same registers as the latest AVX does. And the low level compute operations actually done by the execution units are the same. You need to understand that each instruction is actually translated to one or more micro operation by the decoder, they are not direct execution control data.

However there are some old no longer used features in x86 CPUs that do complicate the design somewhat. And there are instructions connected to those features. But that’s really not the instructions themselves using the die area. Intel’s x86s standard proposes to remove for example the middle privilege level rings and call gates from the CPUs. As well as some no longer relevant memory access modes.

No.

CPU performance is always limited by some reason. In laptops it will almost always be heat, which is the same as power consumption. There is no reason not to drive the CPU as hard as it can on a given cooling and battery solution. You bought a high spec lenovo laptop which is probably configured to run as fast as possible. Did you even try to setup it otherwise?

Typical idle use (like typing reddit) power for my old lenovo intel laptop is about 5W which results in about 8 hour battery life with this small old battery. I don’t think I ever hear the fans of the laptop unless I am running a compiler or something. I could setup it so that I would not hear the fans even then but that would cost some performance.

20A should be manufacturing ready early next year if they are going to launch arrow lake on it in 2024. I suspect it’s already basically done and they are expanding capacity now.

TSMC has had better low power performance but the difference isn’t massive. I suspect the main reason is a) 18A is not ready for mass production until too late and b) limited 20A capacity coming online next year is needed for other products (remember intel has to ship absolute shitload of chips). So why not do it on TSMC if that is possible and the alternative is to delay the products to wait for capacity?

Edit: i’m not sure if 20A was again a more limited early version designed primarily for intel desktop products and 18A was supposed to be the long term version of the node with wider array of capabilities?

Lunar lake is essentially meant for the low power laptop market. By capability comparisons it seems to directly target Apple M3 chip.

So small laptops.

Product naming is fun.

It’s also fun to start looking for their linux driver for the first time. You find something called i915 and wonder what it has to do with your gpu.

And that’s because intel 915 chipset from two decades ago was the last one to receive its very own graphics driver, after that they have just updated the existing one to include new devices. The driver was never renamed.

Looking at this, I don’t think the APO will be a long lived feature. The work it does (or rather the work intel people have apparently done per game and cpu basis) should be done by the game developers, dividing their workloads properly to latency sensitive stuff and background stuff. I think in the future there will be less and less need for this feature.

15th gen desktop is going to bring modest performance boost (think 5-10% single threaded) but significant efficiency boost. That’s the main rumor.

They redesigned it completely as intel phased out Xeon phi. It’s still a couple of years delayed though.

What he proposed isn’t really stacked cache. At least not in the sense AMD does it. It should probably be thought as a separate cache die.

No.

Firstly, intel4 has been shipping for some time now so that’s where they currently are. They say they are in track for intel20A long before tsmc has 2nm. But more importantly, the nanometer number is entirely made up. Nothing in the chips is 4nm. The only thing you should be interested in is performance. If AMD performs better for the price buy AMD, if intel performs better for the price buy intel.

Then you realize few customers actually buy either of those. Most sales are in the ~30ish core products. And it starts to make sense why AMD doesn’t just capture the entire market but is still sitting at 25% even though the 96 core chip is so great.