Tom SundayAs we all know the major disadvantage of buying a pre-made PC is the cost.

Well, that used to be somewhat true, or perhaps more accurately that you got cheaper OEM rigs built with crappy motherboards / power supplies and other components with various things 'nerfed' by design from the OEM to limit RMAs.

What I see right now though, is that it's significantly cheaper and far easier to buy a prebuilt, whether that comes from big OEMs like Dell / HP etc or smaller SI's like CyberPowerPC or iBuyPower. The SI's tend to give better prices, but the OEMs have better warranties.

This is new though, I built my own back in June for less than a prebuilt of similar spec and quality. I've tried recently to spec out a prebuilt and can't really beat SI / OEM prices right now, at least not with one of the new GPUs.

Tom SundaySounds to me like another grab or invasion into the desktop DIY market? I had a hard day...so I be better off to play 'Metro Exodus' for a while to flush my mind. Perhaps Mom has a leftover piece of Cherry Pie upstairs in the kitchen for yours truly?

Laptops accelerated from 65% of the market in 2019 to 75% in 2020. I think the effect is that fewer resources are pushed into the desktop space in general from all the players. Notice how easy it is right now to get a laptop with a 3060 or 3070, and Tiger Lake H.

From a supplier's standpoint, it's a lot easier and cheaper for a supplier to send all their kit to a few big OEMs. Sign contract with OEM - we will supply you with 50,000 video cards per month delivered here by Xth of the month. Easy and straight to the point, no ambiguity as to what needs to be done. But to that same supplier the DIY market looks more like, keeping inventory on hand in a warehouse to supply 100 different retailers with potentially 10,000 different locations with 5-10 GPUs on demand.

InVasManiThe Atom cores are fine for a lot of the OS services a bit of compression and decompression along with rapid storage and other odds and ends and light program tasks and usage. Basically you've got a chip to offload all of that while you have another chip that can have close to 100% CPU usage dedicated to the game itself. I don't see it being perfect initially, but I'll bet you Intel improves it and the second iteration will improve quite a bit further. I don't think content creators are who Intel's trying to sell them to they want to compete with AM4 Ryzen the best they can til a more proper node shrink puts them in a better position to really compete more closely evenly with them again.


So a guy from Fidelity Investments makes lots of money...

InVasManiThe Atom cores are fine for a lot of the OS services a bit of compression and decompression along with rapid storage and other odds and ends and light program tasks and usage. Basically you've got a chip to offload all of that while you have another chip that can have close to 100% CPU usage dedicated to the game itself. I don't see it being perfect initially, but I'll bet you Intel improves it and the second iteration will improve quite a bit further. I don't think content creators are who Intel's trying to sell them to they want to compete with AM4 Ryzen the best they can til a more proper node shrink puts them in a better position to really compete more closely evenly with them again.


So a guy from Fidelity Investments makes lots of money...

Well Harry is a simple "Acount Manager." Not exactly high on the Fidelity exec-ladder. But at 36 still living with Mom with unlimited rights to the refrigerator and free Internet. But when with him I get to live vicariously besides having a good time. I even treat for lunch after the show at Jack in the Box. Indeed there are a lot of Harry's out there and absolutely loved by the show-dealers doing their walk and talk.

qubit@Tom Sunday Dammit, I hope you're wrong about the shift away from DIY PCs...

I was looking a bit into the future...but for the big-three (DELL, Lenovo, HP) the future cannot come early enough. Come to think...DIY will essentially never go away in its entirety, but it's current method, access and pricing will be challenged. Enthusiasts like us will most likely once the bell tolls having to reach deeper into their pockets in securing the real 'premium hardware' for their builds. In turn, I simply abhor the very thought in being forced to buy or use sub-standard hardware specifications and then those being served up cold in propiatary fashion.

Tom SundayWell Harry is a simple "Acount Manager." Not exactly high on the Fidelity exec-ladder. But at 36 still living with Mom with unlimited rights to the refrigerator and free Internet. But when with him I get to live vicariously besides having a good time. I even treat for lunch after the show at Jack in the Box. Indeed there are a lot of Harry's out there and absolutely loved by the show-dealers doing their walk and talk.


I was looking a bit into the future...but for the big-three (DELL, Lenovo, HP) the future cannot come early enough. Come to think...DIY will essentially never go away in its entirety, but it's current method, access and pricing will be challenged. Enthusiasts like us will most likely once the bell tolls having to reach deeper into their pockets in securing the real 'premium hardware' for their builds. In turn, I simply abhor the very thought in being forced to buy or use sub-standard hardware specifications and then those being served up cold in propiatary fashion.

Hey this is Harry, do I know you ?

Desperate much!!! Someone in Hollywood should make a movie about Intel and call it (Desperado 2.0)

Why exactly do we need the Gracemont cores on desktop?

Max(IT)Who is going to buy in March a CPU that will be replaced in September ? :kookoo:

Sure AL is going to get whole next gen platform, ddr5, pcie gen5 but not only is going to be an expensive platform some might have stability/compatibility issues and such, since early adopter is always experiencing this kind of stuff.
Some people with 2-3 year+ upgrade cycle might just want to invest to RL with guaranteed system stability/compatibility also cheaper components and wait till next gen platform matured enough also cheaper enough.

PooPipeBoyIntel overclocked their business plan for product releases from 0.5 generations per year to 2 generations per year. That's an impressive 300% boost in performance.

I honestly hope that Alder Lake S is actually not a joke. I'm tired of seeing Intel recycle their own e-waste for years on end, I haven't wanted to buy a consumer Intel processor since Skylake in 2016.

My sentiments exactly. AMD 5000 desktop w/ APU doesn’t seem is going to materialize.
Don’t need 250 watt arc welder either.
Alder Lake will get a shot.

qubit@Tom Sunday Dammit, I hope you're wrong about the shift away from DIY PCs as it would be a total disaster for enthusiasts like us. We can only wait and see.

I want to upgrade my ancient 2700K (see specs) and so far, it looks like it would be worth waiting for Alder Lake in order to get something current that's built on the latest process, not a backport, like Rocket Lake. I don't need that sabotaged by a shift to complete systems only.

I'm in the same boat. My 3770K CPU is still going strong, especially that I play games in ~3K resolution, where the GPU is still the limiting factor.

What's the point of the Rocket Lake then???

Max(IT)Who is going to buy in March a CPU that will be replaced in September ? :kookoo:

September is annoucement which is meaningless. Could easily mean Q4 for laptops, Q1 2022 for OEM and Q2 2022 for desktops with large inventory/full lineup not avaibility till Q3 2022

qubitI want to upgrade my ancient 2700K (see specs) and so far, it looks like it would be worth waiting for Alder Lake in order to get something current that's built on the latest process, not a backport, like Rocket Lake. I don't need that sabotaged by a shift to complete systems only.

I don't think it would be a big deal. 11th gen is going to have PCI 4.0 support and DDR4 (especially with good timings) will be plenty fast for years to come.

I don't see any sabotaging happening

tussinmanI don't think it would be a big deal. 11th gen is going to have PCI 4.0 support and DDR4 (especially with good timings) will be plenty fast for years to come.

I don't see any sabotaging happening

By waiting a few more months, I'll get a much better system. It's totally worth the wait.

I hope you're right about the complete systems sabotage not happening, or us PC enthusiasts are up Schitt Creek.

qubitBy waiting a few more months, I'll get a much better system. It's totally worth the wait.

It's not going to be a few months. September is the annoucement but it's looking like desktop chips in volume isn't happening till Spring 2022. You could already have faster ram, PCI 4.0, and NVME for an entire year before Alder even hits the shelves.

qubitBy waiting a few more months, I'll get a much better system. It's totally worth the wait.

I hope you're right about the complete systems sabotage not happening, or us PC enthusiasts are up Schitt Creek.

Is it better, though? Nobody answered my question from a week ago, so I still do not know the benefit of the big.little approach on desktop.

You are still only getting 8 high performance cores. And I fear the core management might be a complete mess for a while. Like how will apps/games know which cores to utilize? Has it not been proven, that unconventional architectures are not right for desktops? Bulldozer? Even the CCX approach in first Zen CPUs was not great.

10 nm also does not necessarily mean better. Will these CPUs reach high clock speeds? Process maturity is way more important than power consumption in that respect. Remember Broadwell?

Also, DDR5 will be slower and more expensive for the first few years, just as it always is. It will start to make sense when you can get 4800+ modules for the same price you can get 3200 modules now, and those are super cheap. Even 3600 CL16 is cheap.

Rocket Lake actually gets all the bells and whistles. 8 high performance cores with high clock speeds, PCI-E 4.0 for GPU, as well as on-die Gen4 NVMe link. Plus two times faster DMI resulting in more M.2 slots and native USB 20 Gbps support.
This platform literally has everything you will possibly need for 3-4 years. No quirks to work out, no problems, "it just works".

If you need more cores, go for Zen 3. Waiting for Alder Lake seems pointless. Rocket Lake will be the way to go if you want to stick with Intel.

THU31Is it better, though? Nobody answered my question from a week ago, so I still do not know the benefit of the big.little approach on desktop.

You are still only getting 8 high performance cores. And I fear the core management might be a complete mess for a while. Like how will apps/games know which cores to utilize? Has it not been proven, that unconventional architectures are not right for desktops? Bulldozer? Even the CCX approach in first Zen CPUs was not great.

10 nm also does not necessarily mean better. Will these CPUs reach high clock speeds? Process maturity is way more important than power consumption in that respect. Remember Broadwell?

Also, DDR5 will be slower and more expensive for the first few years, just as it always is. It will start to make sense when you can get 4800+ modules for the same price you can get 3200 modules now, and those are super cheap. Even 3600 CL16 is cheap.

Rocket Lake actually gets all the bells and whistles. 8 high performance cores with high clock speeds, PCI-E 4.0 for GPU, as well as on-die Gen4 NVMe link. Plus two times faster DMI resulting in more M.2 slots and native USB 20 Gbps support.
This platform literally has everything you will possibly need for 3-4 years. No quirks to work out, no problems, "it just works".

If you need more cores, go for Zen 3. Waiting for Alder Lake seems pointless. Rocket Lake will be the way to go if you want to stick with Intel.

Sure, RL is very good, but I'd rather see what AL offers, in official reviews, before making the decision. From experience, I will want the latest product.

I have no idea either how the big.little approach helps the desktop and by the sound of it, there's not much documentation on it, hence others don't know either. What I do know, is that Intel wouldn't invest lots of dollars in developing it if there wasn't a benefit.

I am waiting for platform socket 1700 with DDR5. I have waited years, I can wait a bit longer, I hope.

I str

voltageI am waiting for platform socket 1700 with DDR5. I have waited years, I can wait a bit longer, I hope.

I strongly suspect it will be a long wait...

Max(IT)I strongly suspect it will be a long wait...

Not only that I have a feeling that for most general functions (office work + 1440p gaming) there really won't be a noticeable difference between highend DD4 and 1st gen DDR5 yet you'll not only have to wait 1 to 1.5 years but you'll be paying more money.........

CobainUnless you think that paying 360€ for a 5600x is Smart. That's if you can even find one.

Plenty in stock in DE.
Looks pretty good vs 10700k, so, why?

There is a crisis in semiconductor industry.

It will not be easer until 2023:
www.theregister.com/2021/04/16/tsmc_chip_forecast/

THU31Why exactly do we need the Gracemont cores on desktop?

I don't really think we do.
big.LITTLE is better suited for laptops, where power efficiency is king, not desktops, where (unless Intel fails at getting heat out of their die) it doesn't really matter since you can customize your cooler.

Side note, does anyone else see the irony in big.LITTLE?

RealKGBI don't really think we do.
big.LITTLE is better suited for laptops, where power efficiency is king, not desktops, where (unless Intel fails at getting heat out of their die) it doesn't really matter since you can customize your cooler.

At least Intel can say that they've done something truly innovative (that no one asked for). All hail the new FX!

AusWolfAt least Intel can say that they've done something truly innovative (that no one asked for). All hail the new FX!

Intel's RT. It's cool, but not really usable yet.

And then with (maybe) Zen 5 we'll have AMD big.LITTLE chips.
I hope they make some big.LITTLE and some not, so like the Ryzen 5 7500 BL (big.LITTLE), 7500 BLX (big.little X version), and so on.

RealKGBIntel's RT. It's cool, but not really usable yet.

And then with (maybe) Zen 5 we'll have AMD big.LITTLE chips.
I hope they make some big.LITTLE and some not, so like the Ryzen 5 7500 BL (big.LITTLE), 7500 BLX (big.little X version), and so on.

Like you said in your previous post, I don't really see why big.LITTLE on desktop would ever be a good idea. The concept is that you can turn the big cores down (or completely off) under light loads to save power, and use them only when needed. It's useful for phones and laptops, but for desktop PCs, not so much. I'm not saying that you shouldn't think about your bills and try to save on power when you can, but I think even high-end modern CPUs are efficient enough under light loads so that we don't need to bog the system down with small cores just to save a couple Watts in idle.

For an audio workstation I disable ALL boost developments.
We simply need all cores at a stable speed, preferably 4GHz.

Intel lost out from not having a fast enough reply to AMD Zen.
Just adding more cores/heat and fancy marketing slogans isn’t going to work.

At least they admit 2023 will be the year they get back on track.

You guys are missing the reason that with an additional 8 little cores intel marketing now gets to say they have 16 cores on their desktop cpu.

It should actually be helpful in all core workloads as well. Maybe equivalent to 12 big cores with 24 threads?