• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Discussion Intel current and future Lakes & Rapids thread

Page 862 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
A "generational leap" from Intel's previous offerings, or a generational leap that makes them more efficient than AMD? I dont really follow the mobile market, but I would assume AMD is far ahead in power consumption in mobile as well, so a "generational leap" from intel might only reach parity with AMD's current offerings. We will see when (if?) Intel reaches process lead or parity, but I think they need a new core design to get power usage down to the level they need.
It’s a new node with significantly less current leakage. There’s architectural changes that improves efficiency in certain C states as well.

I don’t know how it’ll do against AMD’s mobile offerings but I’d be really surprised if it wasn’t more efficient or at least on par with Phoenix.

Edit: I’m assuming you’re talking about Phoenix, if you’re referencing Dragon Range (7945HX) RPL already has superior battery life compared to their chiplet mobile processors.
 
Sounds like they will be maximizing the LP core usage in the SoC tile. I don't see how else they can reduce power, unless their Thread Director coupled with new as-yet-unreleased scheduler improvements in Win11, has become 99% accurate in predicting which core(s) will complete the task most efficiently, all the while not letting the user sense the slowed down frequency on battery power.
Have you used a Raptor Lake processor? The thread director is already very good. It all works together really well.
 
It will be interesting to see how they market MTL.

Probably as having better battery life than Raptor Lake.

A "generational leap" from Intel's previous offerings, or a generational leap that makes them more efficient than AMD?
Likely the former. Intel has been going downhill in mobile efficiency for awhile now. Meteor Lake should turn that ship around. The real tragedy is that we'll never see anything but 6+8 configs on Intel 4 (at best).
 
Likely the former. Intel has been going downhill in mobile efficiency for awhile now
4th gen was when it all began tumbling down imo. had both 4th gen mbps and windows laptops. it was abysmal. Skylake was a good performer but it also sucked power. we know how the next 4 years went by.
 
While there's no question that MTL will be more efficient, how much will that really matter in battery life metrics? Under any actual load battery life quickly evaporates on both AMD and Intel. And for normal light usage where 10+ hour figures are possible the rest of the system components are of greater importance.

Which is to say that I don't think there can be a 'generational leap' in laptop battery life from anything other than the batteries themselves. Outside of that we're likely stuck in the realm of incremental improvements.
 
amd did discuss they want to be able to shut down parts of their socs when they're not needed but in true amd form I'd rather wait until the 2nd generationof such an attempt before wastng money on a product like that. intel has or had similar future plans. sounds like a buncha hooey to me.
 
In case it was missed. Meteor Lake is apparently in mass production and will launch Q4. Intel 4 production ramping. Intel 3 mass production ready between now and EOY. Intel 20A mass production ready H1 2024. 18A in H2 2024.
 
they will return and amd is no better lol
Only if AMD makes the same mistake as Intel did in the Skylake era - go to sleep. But knowing Lisa Su and her team, that won't happen. With AMD's acquisition of Xilinx, which is one of the largest acquisitions in that industry, they confirm their ambitions, and now they have all the necessary tools to remain competitive. The era of Intel's dominance is over, Intel will struggle every year and will be forced to work at full capacity if they want to survive.
 
Intel will struggle every year and will be forced to work at full capacity if they want to survive.
Yeah. Their only recourse is to keep designing new instruction set additions to keep x86 relevant as long as they can and make AMD play catch-up on instruction set parity and thus, behind them. AMD doesn't want to do that coz other than x86-64, they have never had success in getting the market to adopt their instruction set enhancements.

It's very telling that Intel pulled AVX-512 right before Zen 4, thus preventing its wider adoption among developers. They can keep giving the excuse of E-cores not supporting it but the fact that it was available at Alder Lake launch and pulled due to an executive order from the highest echelons of the company (Pat? If so, most idiotic move ever), makes it seem like they wanted to avoid direct comparisons and AVX-512 benchmark shootouts because Intel was going to come out looking REALLY bad, especially since they invented it and then got one-upped by their competitor. That's the worst sort of embarrassment that they strategically avoided in the client space.
 
While there's no question that MTL will be more efficient, how much will that really matter in battery life metrics? Under any actual load battery life quickly evaporates on both AMD and Intel. And for normal light usage where 10+ hour figures are possible the rest of the system components are of greater importance.

Which is to say that I don't think there can be a 'generational leap' in laptop battery life from anything other than the batteries themselves. Outside of that we're likely stuck in the realm of incremental improvements.
If the rumors are true, MTL is a brand new design capable of shutting down the CPU tile, the GPU tile & the I/O tile at will. You read that right.. even the CPU tile. And can idle using the new LP E-cores in the SoC tile. And the GPU tile can idle more often using the new ADM L3 (directly linked to it) as a frame buffer, esp. when there isn't much drawing happening.

All these combined can give amazing power savings under normal usage. Maybe even full day battery life like a mac. Might even comfortably beat AMD mobile cpus by a mile. Thats probably the reason they're even rebranding MTL as Ultra. But all this is just speculation as of now (based on leaks). We'll know the full picture only when we actually see a real MTL cpu in action.
 
Last edited:
Only if AMD makes the same mistake as Intel did in the Skylake era - go to sleep. But knowing Lisa Su and her team, that won't happen. With AMD's acquisition of Xilinx, which is one of the largest acquisitions in that industry, they confirm their ambitions, and now they have all the necessary tools to remain competitive. The era of Intel's dominance is over, Intel will struggle every year and will be forced to work at full capacity if they want to survive.
In your ryzen dreams lol .. intel will always dominate as they are doing now
 
In your ryzen dreams lol .. intel will always dominate as they are doing now
LOL funny how dominating means being utterly destroyed in server performance across the board, losing in gaming workloads on desktop, and needing 2x more power to basically match the MT performance of 16C AMD part. The only place they are in front is ST and that is thanks to crazy high boost clocks of 6Ghz on their highest end desktop SKU.

Looking at 2024 parts, even the intel leakers on Twitter are saying that AMD is focusing on ST domination with Zen 5, so Arrow Lake better bring +30% IPC gain versus Raptor Lake with similar clocks (which is basically mission impossible). Fun times ahead!
 
In case it was missed. Meteor Lake is apparently in mass production and will launch Q4. Intel 4 production ramping. Intel 3 mass production ready between now and EOY. Intel 20A mass production ready H1 2024. 18A in H2 2024.
It was mentioned a while ago but certain people went into semantics of intel's own guidelines and comments by "industry experts" of how it'll be late. cheers
 
In case it was missed. Meteor Lake is apparently in mass production and will launch Q4. Intel 4 production ramping. Intel 3 mass production ready between now and EOY. Intel 20A mass production ready H1 2024. 18A in H2 2024.
What's worth mentioning is Meteorlake in 2H2023 would be minor shipment only, mass shipment would happen in 1H2024.
 
AMD has the strongest product, and most new server clients tend to choose AMD. Intel has old clients due to decades of dominance, but they are slowly losing them. One of the ways they are trying to retain them is by giving them their hardware for free or at significantly reduced prices, which explains those significant financial losses. Intel is powerless; they've lost their dominance, and the only thing they can hold onto is this: it's better to give their product away for free and keep the client rather than handing them over to AMD. So, AMD dominates, and as I mentioned, the market share alone doesn't say much—the trend is that clients are switching to AMD.
 
AMD has the strongest product, and most new server clients tend to choose AMD. Intel has old clients due to decades of dominance, but they are slowly losing them. One of the ways they are trying to retain them is by giving them their hardware for free or at significantly reduced prices, which explains those significant financial losses. Intel is powerless; they've lost their dominance, and the only thing they can hold onto is this: it's better to give their product away for free and keep the client rather than handing them over to AMD. So, AMD dominates, and as I mentioned, the market share alone doesn't say much—the trend is that clients are switching to AMD.
It is exactly that type of thought that got Intel into trouble.

1) You are misjudging your opponent. This isn't an Intel vs. AMD war. They are on the same x86 team. The actual war is Intel+AMD vs. ARM. You could throw Nvidia in there too as right now the money and market share is going to ARM + Nvidia. The more you think this is an AMD vs Intel battle, the more AMD will lose the war against ARM.

2) Soon AMD will have access to inferior production equipment. That doesn't mean that AMD will do poorly. But, look at what happened to Intel when they stuck with DUV lithography too long.

Intel got cocky thinking that they could keep succeeding with the current trends at the time. I hope AMD doesn't hear your words and do the same.
 
Last edited:
It is exactly that type of thought that got Intel into trouble.

1) You are misjudging your opponent. This isn't an Intel vs. AMD war. They are on the same x86 team. The actual war is Intel+AMD vs. ARM. You could through Nvidia in there too as right now the money and market share is going to ARM + Nvidia. The more you think this is an AMD vs Intel battle, the more AMD will lose the war against ARM.

2) Soon AMD will have access to inferior production equipment. That doesn't mean that AMD will do poorly. But, look at what happened to Intel when they stuck with DUV lithography too long.

Intel got cocky thinking that they could keep succeeding with the current trends at the time. I hope AMD doesn't hear your words and do the same.
…and RISC-V.

There are some real interesting RISC-V projects coming down the pipe, including a chip made on Intel 4/3.

At least one RISC-V chip coming is offering desktop class performance comparable to an 8 core gracemont cluster (it isn’t a desktop class part, however, it is destined for server use)

Top end RISC-V chips are currently around the Raspberry Pi 4 in terms of performance.

Between ARM/RISC-V and x86 64 emulators, things will be looking rather spicy in the future.
 
What's worth mentioning is Meteorlake in 2H2023 would be minor shipment only, mass shipment would happen in 1H2024.
yes minor shipment would be to specific odms like dell and hp who will use it to spur holiday sales into an otherwise crappy q1.
 
2) Soon AMD will have access to inferior production equipment. That doesn't mean that AMD will do poorly. But, look at what happened to Intel when they stuck with DUV lithography too long.

Intel got cocky thinking that they could keep succeeding with the current trends at the time. I hope AMD doesn't hear your words and do the same.
This banks on Intel's first attempts at mixed ip tiles not becoming a financial burden in the long run and them being able to make sense of backside power delivery and gaafet. The same cocky attitude that got intel in trouble is resurfacing for them and even their die hard fans. same goes for amd fans. It's exhausting. fans are exhausting..
 
You still don't realize that he's just trolling you, lol. I know him from other places; he's not an Intel fan at all, just a troll. That's his entertainment.
Thanks for the tip. I've never blocked anyone on these forums but perhaps it's time to start.
 
Back
Top