Why is the i7 4870 weaker than the i7 4810?

It's frequency is lower by 100mhz.

This doesn't make sense, it's 1 year older from a recent refresh and has a higher number and is more expensive and has the same power usage.

What is this backwards logic?

 

never even knew a 4870MQ existed, I'd be interested to know as well

 

That's i7 4870 HQ, the version with GT3e. Its iGPU is better, the CPU part is not.

Intel Core i7 4870HQ Notebook Processor - NotebookCheck.net Tech

 

I find the iGPU annoying in terms how it misleads the ordering of "how good is a CPU", since its pretty much irrelevant whn u have a discrete GPU. Or am i wrong about this?

 

It depends. When you do have a faster dGPU, maybe. But when you're not into the largest dGPU money can buy but do need some GPU power, an iGPU that's just good enough would increase the over all efficiency considerably (for both space in the device and energy consumption). If it happens to have just the right amount of power for your needs, a more compact device with better battery life is definitely not bad. There's also the special case of using the iGPU for viewport and the dGPU headlessly in parallel, in which performance of both GPUs would matter.

Building CPU cores and an iGPU on the same die would open up the opportunity for HSA as well. Unfortunately AMD CPU cores are too weak to be relevant for now and Intel is not buying this idea for whatever reason. The Intel iGPU is already fast enough for plenty of stuff, and if Intel made something similar to HSA/HUMA games would have way more computing power for doing funny stuff with physics/AI while the dGPU is busy with the rendering.

 

Welcome to processor naming 101. Make up random confusing names and numbers that most consumers won't bother to check. Force those who are actually interested to turn to third parties to decipher the hieroglyphics.

 

It applies to dGPUs as well, or pretty much everything you can buy from the modern market.

 

When something is numbered higher, it's usually better/faster, this is very misleading.

If I were in the market looking for a laptop and saw one with an 4870 and another with a 4810 I will definitely and blindly go for the 4870 thinking it's faster!

 

Intel, Nvidia, AMD, BMW, Boeing, etc. What is the world coming to?

 

More like "What is the world not coming to?"

 

Just because the 4870HQ is clocked 100mhz slower doesn't mean it'll perform slower... the 128mb of eDRAM should increase CPU performance over a non-eDRAM part.

 

messed up, didn't research before I posted. lesson learnt

 

This is done for a reason. The average joe will look at the 4870 compared with the 4810, and see the lower price, and convince themselves into an impulse buy simply because it bust be better than the 4810 because it's a larger number! The enthusiast will do research, like expected, and make an informed decision. It's all market tactics to make resisting an impulse buy that much harder.

It's like how costco shuffles things around all the time, so you buy things you weren't intending as you run across it.

 

I would argue that while the general naming scheme is misleading this specific one is not that bad. You only lose 100m clock on the CPU front but get a significantly better iGPU. Even if you don't need that, you don't lose much.

I was under the impression that the L4 cache is only used for the iGPU? Would be very happy to be proven wrong though.

Edit: Oops. Some simple Googling revealed it is available on the CPU and is effective. So in this case the bigger number is better (at least when you're not purely bounded by the core).

 

higher number usually mean higher price, and pretty much it.