I have a brand new Core i9-14900F in an Alienware Aurora R16. I also have a brand new M4 Max in a 16” MacBook Pro.
The M4 is ~40% faster on almost every test I’ve run (I’ll be sharing raw data in a blog soon). The M4 is faster on battery than the i9 is sucking juice from the outlet.
I feel like there are issues more fundamental than supply shortages. A wide gap has opened up in performance and an even wider gap in performance/power.
The 14900 does not reflect Intel’s radically more efficient upcoming 18A and 14A nodes which is what much of the hype around the company’s revival is hoped on. Time will tell when products based on 18A come out in a few months.
Here is the issue. Apple is two years ahead. They've been two years ahead for a while, and they will be two years ahead going forward a few years or forever.
Can you elaborate on this? I'm curious to hear more details about this.
For example, does a Asus ROG laptop match M chips from Apple? For example, I see Asus ROG g16 indicates a 13th gen Intel chip. Am I wrong to assume these are the same thing and don't suffer the same problems?
I don’t know. I feel like we could also make claims about Apple hardware that isn’t out yet. “M4 does not represent Apple’s radically different M6 processor.”
Intel may get faster but the competition should also be expected to get faster.
Intel have a track record of over promising and under delivering. I wouldn’t be surprised if this was what pushed Apple into creating the M series processors.
I used hamstrung at work recently and I reflected that it is somewhat old fashioned English word. I asked a friend afterwards and they were unfamiliar with the term.
In doing investigation around the word, I discovered its origins are in the brutal history of the transatlantic slave trade, where captured runaway slaves would have their hamstrings severed to prevent them from ever running again.
While the brutal practice of hamstringing runaway slaves was even codified into french colonial law, the practice and I think even the term predates this. The practice of hamstringing enemy horses is even mentioned in the Old Testament.
I predict we will see a death of intel similar to the death of Chrysler in the next decade.
Too little too late has been done to save it. it ran rudderless for 30 years under the hand of marketers and grifters who used stock buybacks and deception to ensure it looked strong in the press. It pioneered things like gaming compiler results to achieve benchmark supremacy. it squandered its potential at the helm of a leadership that cared more about profit than innovation.
Perhaps it will sell to Texas Instruments or motorola but these arent the cherished powerhouses of industry our octogenarian congress reminisces they were. Motorola spends its days focused on niche telecom like apco p25 and IoT nannyware like the snitch puck https://infocondb.org/con/def-con/def-con-33/unmasking-the-s.... it develops very little of the SoC or chips it uses.
the government needs intel, but i suspect will in administrations to come grow increasingly weary and frustrated with its morass of managerial bloat, bureaucratic stagnation and inability to evolve the business model.
> I predict we will see a death of intel similar to the death of Chrysler in the next decade.
They have a ways to go.
Chrysler merged with Mercedes (Daimler). Then they were sold to Cereberus. Then filed for bankruptcy. Became 'new' Chrysler. Then bought by Fiat. Merge with Puegot.
I actually don’t know if it’s good or bad for Intel.
I could see an embargo or tariffs on TSMC chips… but is Intel better with competition or worse? Do other countries care as much as the US? They may buy TSMC anyways.
Supply side constraints are typically bullish and good for a business. I was buying Intel stock hand over fist at peak media fear mongering as I knew their demise was greatly exaggerated. $INTC is up 90+% YTD.
I have a brand new Core i9-14900F in an Alienware Aurora R16. I also have a brand new M4 Max in a 16” MacBook Pro.
The M4 is ~40% faster on almost every test I’ve run (I’ll be sharing raw data in a blog soon). The M4 is faster on battery than the i9 is sucking juice from the outlet.
I feel like there are issues more fundamental than supply shortages. A wide gap has opened up in performance and an even wider gap in performance/power.
I’d like to know Intel’s answer to that.
The 14900 does not reflect Intel’s radically more efficient upcoming 18A and 14A nodes which is what much of the hype around the company’s revival is hoped on. Time will tell when products based on 18A come out in a few months.
Here is the issue. Apple is two years ahead. They've been two years ahead for a while, and they will be two years ahead going forward a few years or forever.
If we pretend Strix Halo doesn't exist, sure.
I get your point but Strix Halo isn’t Intel. Maybe another company can close the gap but it still doesn’t bode well for Intel.
> I get your point but Strix Halo isn’t Intel.
But of course, Apple isn't Apple, nor is AMD AMD - they're both Taiwan's TSMC.
It isn’t clear that TSMC is all the special sauce here. The Apple chips are uniquely good in my experience so far.
Maybe another chip has similar power/performance but:
1. I haven’t seen real-world examples showing that
2. It might not run the desktop workloads I care about
Maybe some other TSMC-based device is out there that suits my needs but if it is, it at least has piss poor advertising.
Qualcomm and Apple are neck to neck, inching ahead of each other with each new design - both use TSMC for production.
Can you elaborate on this? I'm curious to hear more details about this.
For example, does a Asus ROG laptop match M chips from Apple? For example, I see Asus ROG g16 indicates a 13th gen Intel chip. Am I wrong to assume these are the same thing and don't suffer the same problems?
Any links I should review?
That matches performances, but sadly not efficiency.
I don’t know. I feel like we could also make claims about Apple hardware that isn’t out yet. “M4 does not represent Apple’s radically different M6 processor.”
Intel may get faster but the competition should also be expected to get faster.
Intel have a track record of over promising and under delivering. I wouldn’t be surprised if this was what pushed Apple into creating the M series processors.
Memory bandwidth on that Apple chip is like x3 than that of the Intel. No surprise it’s faster.
Yeah that’s right. It’s faster.
In true off-topic HN style...
I used hamstrung at work recently and I reflected that it is somewhat old fashioned English word. I asked a friend afterwards and they were unfamiliar with the term.
In doing investigation around the word, I discovered its origins are in the brutal history of the transatlantic slave trade, where captured runaway slaves would have their hamstrings severed to prevent them from ever running again.
While the brutal practice of hamstringing runaway slaves was even codified into french colonial law, the practice and I think even the term predates this. The practice of hamstringing enemy horses is even mentioned in the Old Testament.
I predict we will see a death of intel similar to the death of Chrysler in the next decade.
Too little too late has been done to save it. it ran rudderless for 30 years under the hand of marketers and grifters who used stock buybacks and deception to ensure it looked strong in the press. It pioneered things like gaming compiler results to achieve benchmark supremacy. it squandered its potential at the helm of a leadership that cared more about profit than innovation.
Perhaps it will sell to Texas Instruments or motorola but these arent the cherished powerhouses of industry our octogenarian congress reminisces they were. Motorola spends its days focused on niche telecom like apco p25 and IoT nannyware like the snitch puck https://infocondb.org/con/def-con/def-con-33/unmasking-the-s.... it develops very little of the SoC or chips it uses.
the government needs intel, but i suspect will in administrations to come grow increasingly weary and frustrated with its morass of managerial bloat, bureaucratic stagnation and inability to evolve the business model.
> I predict we will see a death of intel similar to the death of Chrysler in the next decade.
They have a ways to go.
Chrysler merged with Mercedes (Daimler). Then they were sold to Cereberus. Then filed for bankruptcy. Became 'new' Chrysler. Then bought by Fiat. Merge with Puegot.
What happens in your scenario if China invades Taiwan and TSMC is taken out of the picture?
I actually don’t know if it’s good or bad for Intel.
I could see an embargo or tariffs on TSMC chips… but is Intel better with competition or worse? Do other countries care as much as the US? They may buy TSMC anyways.
You think Taiwan will be exporting chips during a hot war?
I think you replied to the wrong comment. I didn’t say anything about that nor did I mean to imply it.
> They may buy TSMC anyways.
And how would they do that?
They should fab carbon-based chips to eliminate supply chain limits, decrease resistivity, and reduce thermal waste.
CNT (Carbon Nanotubes) on rGO (reduced Graphene Oxide) wafers should work due to the difference in work functions between each form of carbon.
Semiconductor fabrication with (SiC) Silicon Carbide is already demonstrated.
Carbon epoxide (C_n H_2n O_n) would probably also be a sufficient substrate for electronic computing.
/?hnlog graphene, out of graphene :
- "Ask HN: How much would it cost to build a RISC CPU out of carbon?" (2024) https://news.ycombinator.com/item?id=41153490
Supply side constraints are typically bullish and good for a business. I was buying Intel stock hand over fist at peak media fear mongering as I knew their demise was greatly exaggerated. $INTC is up 90+% YTD.