*Official Benchmark Thread* - Post it here or it didn't happen :D

[Papusan]

On 5/26/2022 at 9:12 PM, Papusan said:

Look at the beauty 🙂 An old Point of View GTX 580.

 

Here's a few pics of the old paste and pressure testing of the Gtx 580. A huge old Nvidia heatspreader needs cheapest possible thermal paste. And Colgate toothpaste fits the bill 🙂 The owner said the card was repasted a few months ago. I opened it up to see and can't see he succeded very much in that. A bare die as they use nowadays is a lot better than this huge heatspreader. Not even sure if it's nickel plated copper as Intel do for their IHS. 

 

 

Cleaned off the mess.

 

 

 

Btw. For those that haven't seen it. Clevo is soon out with a new High End gaming laptop  Clevo X270

 

God have mercy with the buyers, if this is the successor to  X170 (I don't even know what I would brand it - Joke or a Turd).

 

 

[Papusan]

ASRock Z690 Taichi Razer Edition Review: Less Zen, More RGBs Z690 Taichi Razer Edition sports all of the Taichi's features and specs with more RGBs and Razer Chroma integration for $620.

 

ASRock have sold their soul to Razer. All to catch/grab the gamer kids/those that value bloatware over awesome. I have never understood why people is willing to pay more for “no added value”.

I will never buy Razer again. 

 

Here's another failure if this results will be the fasit (yep, still rumors). 

Intel Core i9-13900K Raptor Lake Desktop CPU Could End Up 15% Faster Than Core i9-12900K & 35% Faster Than AMD Ryzen 9 5950X In Single-Threaded Performance. 

[Papusan]

14 hours ago, Papusan said:

Cleaned off the mess.

After I cleaned the heat spreader and the copper base with isopropyl/alcohol, everything looked fine. But the vision can be deceiving. A round with Flitz polish shows that a good metal polish will find more dirt. Good quality isopropyl alcohol just show its not enough for the very best cleaning. Aka you'll need, 1 round with  isopropyl/alcohol, then metal polish and then do the last finish with isopropyl alcohol.

 

[Mr. Fox]

1 hour ago, Papusan said:

After I cleaned the heat spreader and the copper base with isopropyl/alcohol, everything looked fine. But the vision can be deceiving. A round with Flitz polish shows that a good metal polish will find more dirt. Good quality isopropyl alcohol just show its not enough for the very best cleaning. Aka you'll need, 1 round with  isopropyl/alcohol, then metal polish and then do the last finish with isopropyl alcohol.

 

Looking good. Did all GTX 580 have a vapor chamber, or only that model?

[Papusan]

2 hours ago, Mr. Fox said:

Looking good. Did all GTX 580 have a vapor chamber, or only that model?

Not sure if all the different brands did it. But it depends if they followed nvidias guidelines (cooling reference design) or not.  What they use, depending on the  different model SKUs in same way as they do nowadays.

GeForce GTX 580 Cooling Solution Detailed to Public

NVIDIA GeForce GTX 580 GPU Features Detailed At PDXLAN

Edit.

http://ixbtlabs.com/articles3/video/gf110-6-p1.html

 

Here's one very old variant of Linus 🙂

 

 

[Mr. Fox]

This was an interesting and informative video. Especially the "unexpected results" (15:56) which amounts to robbing Peter to pay Paul... good to know. Probably would affect an air cooled GPU more dramatically than a water cooled GPU.

 

The Gelid Solutions GP-Ultimate and Nab Cooling pads seem to be the same product with different branding and they not only performed the best, but were among the most affordable.

 

 

[Tenoroon]

18 minutes ago, Mr. Fox said:

This was an interesting and informative video. Especially the "unexpected results" (15:56) which amounts to robbing Peter to pay Paul... good to know. Probably would affect an air cooled GPU more dramatically than a water cooled GPU.

 

The Gelid Solutions GP-Ultimate and Nab Cooling pads seem to be the same product with different branding and they not only performed the best, but were among the most affordable.

 

 

I tried the NAB pads on my 1070 when I first put it into the P870, and they were horrible.

I don't mean they are bad in the sense of thermals, but they were very hard, crumbly, and annoying to cut. They are firm and don't really squish under weird mounting pressure. Not ideal for Laptops.

The thermals for everything seemed fine, but I wouldn't recommend using them for laptops with weird/inconsistent gaps and mounting pressure. They seem to be fine for desktop GPUs though, where the gap/pressure is consistent.

I've been looking for good pads for the 1070 since the VRM area of the card is bent downwards a bit, and I'm worried about bending it more lol. I've heard the Arctic pads are squishy and good under pressure, but I haven't bought any yet. I'd like to use K5 Pro, but I've heard conflicting reports of its effectiveness and have heard from Fox that it's a pain in the ass to clean up.

[Mr. Fox]

2 hours ago, Tenoroon said:

I tried the NAB pads on my 1070 when I first put it into the P870, and they were horrible.

I don't mean they are bad in the sense of thermals, but they were very hard, crumbly, and annoying to cut. They are firm and don't really squish under weird mounting pressure. Not ideal for Laptops.

The thermals for everything seemed fine, but I wouldn't recommend using them for laptops with weird/inconsistent gaps and mounting pressure. They seem to be fine for desktop GPUs though, where the gap/pressure is consistent.

I've been looking for good pads for the 1070 since the VRM area of the card is bent downwards a bit, and I'm worried about bending it more lol. I've heard the Arctic pads are squishy and good under pressure, but I haven't bought any yet. I'd like to use K5 Pro, but I've heard conflicting reports of its effectiveness and have heard from Fox that it's a pain in the ass to clean up.

Your experience is consistent with mine. Pads with very high thermal conductivity ratings that are dense and hard generally suck on laptops. They are better quality and more effective, but you lose the benefit of that when they don't make proper contact and decrease die contact pressure. Most laptops do not have heat sinks that are capable of creating enough mounting pressure to crush the dense pads far enough to conform with uneven surfaces.

 

I use K5 Pro on my laptops now instead of pads. Yes, it is a massive pain in the butt to clean up, but it's not like that is a recurring ritual, so it is not a big deal in the grand scheme of things. (There is no reason to clean it all off completely and have everything looking immaculate unless you are going back to using thermal pads later.) Other than K5 Pro, the extra soft and squishy thermal pads are often best for laptops with crappy fitting components. They easily conform to uneven surfaces and warped heat sinks with contact surfaces that are not precise. That is also the same reason I use K5 Pro. It fills the gaps no matter how uneven they are and results in fewer issues with contact at the die because it offers no resistance to compression. Because it is a dense paste, this will result in the least amount of space between contact surfaces because it will compress to a thickness that is less than a 0.5mm pad. So, die contact is never impeded by pads that are too thick using K5 Pro.

[jaybee83]

1 hour ago, Mr. Fox said:

This was an interesting and informative video. Especially the "unexpected results" (15:56) which amounts to robbing Peter to pay Paul... good to know. Probably would affect an air cooled GPU more dramatically than a water cooled GPU.

 

The Gelid Solutions GP-Ultimate and Nab Cooling pads seem to be the same product with different branding and they not only performed the best, but were among the most affordable.

 

 

nice, ive had my eyes on the gelid gp ultimate pads for a while now but recently also discovered the cooler master thermal pad pro. good thing about those is that pricing is competitive and thermal conductivity is also given for higher thicknesses (1mm - 15.3 / 2mm - 15.3 / 3mm - 13.3). too bad they didnt include those in their testing!

[Mr. Fox]

It took a while, but I finally got 6800 CL30 dialed in as a daily driver memory overclock.

Some interesting discoveries in the process...

• Setting tRFC and tRFCPB too tight improves read speed, but degrades write, copy and latency and the best overall performance and latency are achieved leaving both of them set to Auto and letting the BIOS train the values.
• Likewise, tRCD and tRP set too tight produced errors. Moving from 38 to 39 reduced errors, moving to 40 eliminated them.
• CPU VDDQ and CPU VDD2 and DRAM VPP work best if left on Auto. Setting them manually to the values programmed by the BIOS when set on Auto causes inconsistency in stability.
• Setting tWR and tWR_MR too tight causes inconsistent results, while leaving them on Auto and letting the BIOS train the values had no effect on performance and eliminated errors in testing.
• Setting tWTR and tWTR_L at 14 and 28, respectively, displays as 8 and 19 using Dragon Ball. Asus MemTweakIt doesn't even read those two values and it does not accurately read the tWR value. Using either software tool can result in erroneous interpretation due to the BIOS values not being displayed accurately for a few settings. Most of the values displayed by Dragon Ball match the BIOS settings.

 

 

[Papusan]

3 hours ago, Mr. Fox said:

This was an interesting and informative video. 

 

The Gelid Solutions GP-Ultimate and Nab Cooling pads seem to be the same product with different branding and they not only performed the best, but were among the most affordable.

 

 

I prefer the cheaper variants from Gelid as the Gelid GP-Extreme pads over the Ultimate. They offer lower Thermal conductivity (12 W/mK) but is softer and more forgiven in use (and cheaper). Higher Thermal conductivity just means the pads will be harder/more stiff and more difficult to use if you struggle with variable hight differences. The end results can often be even worse than lower specked Thermal conductivity pads. No guaranteee that all will get exactly same good results from same type and brand thermal pads.

 

 

------------------------------------------------------------------

 

I wonder what this card will cost if it will be a reality (only rumors).

 

NVIDIA’s Ultimate Ada Lovelace GPU Could Be A Titan Class Behemoth: Rumored To Feature 48 GB & 24 Gbps Memory, Full AD102 GPU, Dual 16-Pin Connectors For Up To 900W

3 hours ago, Tenoroon said:

I tried the NAB pads on my 1070 when I first put it into the P870, and they were horrible.

I don't mean they are bad in the sense of thermals, but they were very hard, crumbly, and annoying to cut. They are firm and don't really squish under weird mounting pressure. Not ideal for Laptops.

The thermals for everything seemed fine, but I wouldn't recommend using them for laptops with weird/inconsistent gaps and mounting pressure. They seem to be fine for desktop GPUs though, where the gap/pressure is consistent.

I've been looking for good pads for the 1070 since the VRM area of the card is bent downwards a bit, and I'm worried about bending it more lol. I've heard the Arctic pads are squishy and good under pressure, but I haven't bought any yet. I'd like to use K5 Pro, but I've heard conflicting reports of its effectiveness and have heard from Fox that it's a pain in the ass to clean up.

 

[Tenoroon]

5 hours ago, Mr. Fox said:

Your experience is consistent with mine. Pads with very high thermal conductivity ratings that are dense and hard generally suck on laptops. They are better quality and more effective, but you lose the benefit of that when they don't make proper contact and decrease die contact pressure. Most laptops do not have heat sinks that are capable of creating enough mounting pressure to crush the dense pads far enough to conform with uneven surfaces.

 

I use K5 Pro on my laptops now instead of pads. Yes, it is a massive pain in the butt to clean up, but it's not like that is a recurring ritual, so it is not a big deal in the grand scheme of things. (There is no reason to clean it all off completely and have everything looking immaculate unless you are going back to using thermal pads later.) Other than K5 Pro, the extra soft and squishy thermal pads are often best for laptops with crappy fitting components. They easily conform to uneven surfaces and warped heat sinks with contact surfaces that are not precise. That is also the same reason I use K5 Pro. It fills the gaps no matter how uneven they are and results in fewer issues with contact at the die because it offers no resistance to compression. Because it is a dense paste, this will result in the least amount of space between contact surfaces because it will compress to a thickness that is less than a 0.5mm pad. So, die contact is never impeded by pads that are too thick using K5 Pro.

Would you recommend that I just use K5 Pro then? 

I'm mostly worried about the giant gap from the VRM to the heatsink, its like a 2.5-3.0mm gap, you think enough K5 Pro would be able to cover that?

Also, how well does K5 Pro act when you take the heatsink off of the applied surface? I will be taking my heatsink off many times in order to modify it to cool the 1070 better, and would like to avoid re-applying K5 Pro every time.

[Clamibot]

K5 Pro seems to hold pretty well. I've taken off the heatsink on my Clevo X170 and put it back on multiple times now without reapplying anything. The first time I took off the heatsink, I thought the thermal pads had melted! I didn't know what K5 Pro was until I found out that zTecpc applies it to components that would traditionally have thermal pads on them, such as VRAM chips and VRMs.

 

K5 Pro is gooey, but also pretty viscous. It'll hold for multiple remounts of the heatsink. The same goes for thermal paste. I don't normally bother repasting as I can just reseat the thermal paste that is already there and have the heatsink respread the paste for me as I put it back on.

[electrosoft]

1 hour ago, Tenoroon said:

Would you recommend that I just use K5 Pro then? 

I'm mostly worried about the giant gap from the VRM to the heatsink, its like a 2.5-3.0mm gap, you think enough K5 Pro would be able to cover that?

Also, how well does K5 Pro act when you take the heatsink off of the applied surface? I will be taking my heatsink off many times in order to modify it to cool the 1070 better, and would like to avoid re-applying K5 Pro every time.

 

If I know I'm doing temp test mounts when testing various CPU stuff on my X170SM-G I will leave the K5 Pro as is, but when it is my final remount I will use a thermal paste applicator and clean up the surrounding squeeze out and reapply it. Takes about 10 min tops.

 

 

[Mr. Fox]

4 hours ago, Tenoroon said:

I'm mostly worried about the giant gap from the VRM to the heatsink, its like a 2.5-3.0mm gap, you think enough K5 Pro would be able to cover that?

Go ahead and put a slightly thinner than stock thermal pad there on the VRM and lay a bead of K5 down on top of it. Put the K5 Pro into a big syringe and that makes applying it super easy. You can apply a bead or round glob on each component and let the heat sink compress it. That is SO MUCH faster, easier, cleaner and produces a consistent outcome. Trying to smear is on with a spatula kind of sucks.

 

I will post a picture later for you.

[Mr. Fox]

Brother @Tenoroon...

You should place a thermal pad where indicated in these two spots on a GPU, This is to avoid an electrical short circuit with the heat sink touching live contact points on the VRM resistors. Then use K5 Pro everywhere else a pad used to be.

 

 

 

[electrosoft]

1 hour ago, Mr. Fox said:

Brother @Tenoroon...

You should place a thermal pad where indicated in these two spots on a GPU. Then K5 Pro everywhere else a pad used to be.

 

 

 

 

To add to this, here is what it looks like when you pull the heatsink off after a K5 application. This is before clean up and with a LTX Golden 10900k with a BartX nickel plated IHS installed. It is a wash what sticks to what either the heatsink or the system but like I said before clean up is pretty quick with a TIM spreader.

 

 

[Papusan]

The beauty with Intel. Let you have the chance to upgrade your MB. Maybe they are able to fix their flawed loading mechanism (ILM) design with a new socket? 

 

Intel 14th Gen Core desktop “Meteor Lake” allegedly requires new LGA-2551 socket

It would appear that the rumors stating Intel will keep 3 generations on the same socket are untrue. In the fourth quarter of 2023 when Intel is expected to launch a new desktop series codenamed Meteor Lake, an entirely new socket might be required.

 

And look at this beaty.... Wouldn't it be nice with 32 Baby cores bro @Ashtrix 😀 More cores has to be good (8P + 32-E cores). Intel know exactly how to improve/compete and please the performance enthusiasts.... Very fast P-cores for single threaded tasks and loads of baby cores to help the few P-cores in multi threaded tasks. Jam, Jam. Nice.

 

15th Gen Core Arrow Lake

The Lion Cove/Skymont cores might already debut with Arrow Lake. These new microarchitectures were previously rumored for Lunar Lake. As mentioned in previous leaks, the 15th Gen Core would not see an increase in Performance cores, but Efficient cores count would double over Raptor Lake, now counting 32 in total.

 

 

[Mr. Fox]

Maybe they should release a CPU with 300 Atom cores that run at 1.2GHz and 1 P-core that runs at 10GHz.

[Tenoroon]

6 hours ago, Mr. Fox said:

Brother @Tenoroon...

You should place a thermal pad where indicated in these two spots on a GPU, This is to avoid an electrical short circuit with the heat sink touching live contact points on the VRM resistors. Then use K5 Pro everywhere else a pad used to be.

 

 

 

 

5 hours ago, electrosoft said:

 

To add to this, here is what it looks like when you pull the heatsink off after a K5 application. This is before clean up and with a LTX Golden 10900k with a BartX nickel plated IHS installed. It is a wash what sticks to what either the heatsink or the system but like I said before clean up is pretty quick with a TIM spreader.

 

 

I'll probably go with some K5 then, do you mind telling me where you two purchased yours? I ask because its a less mainstream product, and I would like to avoid purchasing something that is a bad clone.

[Mr. Fox]

25 minutes ago, Tenoroon said:

 

I'll probably go with some K5 then, do you mind telling me where you two purchased yours? I ask because its a less mainstream product, and I would like to avoid purchasing something that is a bad clone.

Amazon. I get the 3 small containers and fill the syringe using the popsicle stick provided with the K5 Pro.

 

https://www.amazon.com/viscous-thermal-replacement-60g-Aspire/dp/B00K04D3UK

I use the 20ml syringe

 

https://www.amazon.com/gp/product/B07JZ2HMJ7?th=1

[Ashtrix]

5 hours ago, Papusan said:

The beauty with Intel. Let you have the chance to upgrade your MB. Maybe they are able to fix their flawed loading mechanism (ILM) design with a new socket? 

 

Intel 14th Gen Core desktop “Meteor Lake” allegedly requires new LGA-2551 socket

It would appear that the rumors stating Intel will keep 3 generations on the same socket are untrue. In the fourth quarter of 2023 when Intel is expected to launch a new desktop series codenamed Meteor Lake, an entirely new socket might be required.

 

And look at this beaty.... Wouldn't it be nice with 32 Baby cores bro @Ashtrix 😀 More cores has to be good (8P + 32-E cores). Intel know exactly how to improve/compete and please the performance enthusiasts.... Very fast P-cores for single threaded tasks and loads of baby cores to help the few P-cores in multi threaded tasks. Jam, Jam. Nice.

 

15th Gen Core Arrow Lake

The Lion Cove/Skymont cores might already debut with Arrow Lake. These new microarchitectures were previously rumored for Lunar Lake. As mentioned in previous leaks, the 15th Gen Core would not see an increase in Performance cores, but Efficient cores count would double over Raptor Lake, now counting 32 in total.

 

 

 

Both AMD and Intel are really massively deep into screwing customers. Shame that Intel is again following same garbage 2 CPU per socket formula as always while AMD is giving so many damn things. I think Intel might fix the ILM in a way it doesn't break the cooler compat with LGA1700 on upcoming Z790 if not they also should be avoided unless one knows how to fix the ILM and handle that DIY tweak.

 

 

Intel is deadlocked into this 8P fixed and will keep on adding small improvements to the P cores while adding more E JUNK. Because their Core is same architecture. I think after 2026 they might do a full new micro architecture. Meanwhile Zen will bulldoze them, god I hope. I hate this E junk nonsense and giving 8P max for that gaming performance joke and milk more on BGA TRASH 🤬

 

Check this information out. I learned some information on AMD's upcoming Zen 4 Chipset. This lane business is always confusing no matter what lol...

 

AMD Zen 4 & Socket AM5 Explained: PCIe Lanes, Chipsets, Connectivity

 

Quote

 

Connectivity from the Processor
Starting with the Zen 4 processor's lanes, all of its PCIe lanes are PCIe 5.0 and there are a total of 28 lanes. The first 16 PCI Express lanes will be used for a single x16 PCIe slot, or they can be split into two x8 slots. AMD's requirements only enforce PCIe 5.0 on the X670E boards, which means PCIe 4.0 will be applicable to lower cost motherboards.

The remaining four PCIe lanes are used for connecting to the chipset. Just to clarify, on the processor side these do support PCI-Express 5.0, the chipset only supports PCIe 4.0, so the link negotiation mechanism will downgrade the link to Gen 4.

 

 

Quote

 

Promontory 21 offers a total of 16 PCI-Express lanes. Four of these are used to connect to the CPU, over a Gen 4 interface, as mentioned before. In the X670/X670E daisy-chained configuration the secondary chipset connects to the primary chipset, it has no direct link to the processor. This means that on the primary chipset another four lanes are used up, leaving eight usable PCIe lanes, whereas the secondary chipset has 12 usable PCIe lanes. Four of the lanes are PCIe 3.0, although these are muxed interfaces with SATA 6 Gbps. This allows the motherboard manufacturers to choose how they want to implement those interfaces and as we've seen, ASRock has gone for eight SATA ports, whereas most other board makers appear to be going for six on their X670 and X670E motherboards.
 

Compared to Intel's Z690 chipset, which has support for a total of 28 PCIe lanes, AMD has clearly decided to scale things back a little bit. In all fairness, Intel doesn't support more than 12 PCIe 4.0 lanes from the Z690 chipset and four of those lanes are shared with SATA 6 Gbps ports. Intel wins by having support for an additional 12 PCIe 3.0 lanes though, but two of those are shared with an Ethernet MAC, something AMD doesn't do, as the company relies on PCIe based Ethernet controllers. It's worth noting that Intel has a wider bus to some of its chipsets, as their CPUs support eight DMI 4.0 lanes. Comparing AMD's B650 chipset with Intel's B660, AMD comes out slightly ahead if high-speed interfaces matter, as the B660 chipset only supports six PCIe 4.0 lanes and eight PCIe 3.0 lanes, although none of its four SATA 6 Gbps ports are shared with PCIe.

 

 

 

Quote

What we obviously don't know at this point is how much of a performance penalty there will be for AMD having two chipsets, especially when it comes to high-speed storage devices. We've seen some concerns about this implementation in virtualized environments and how these devices will appear to the OS in such a case, but we don't share those concerns. We expect the primary chipset to appear as PCIe bridge to the host system, a mechanism that is part of the specification and has been supported and used for many years. These are things we're going to have to wait and see how they play out, but AMD clearly deemed the tradeoffs reasonable enough versus the cost of developing multiple different chipsets.

 

Basically the X670E / X670 Dual Chipset Design has to prove it's Reliability - which AMD didn't show with AM4 as they are still busted with USB issues thanks to GloFo12nm IOD which is JUNK and doesn't work properly with Zen 3 chiplets, and Performance.

 

Also looks like AMD wanted to go somewhat cost effective route (why not when you spend tons of cash on the BS garbage dumpster class junk called BGA trash and give it TSMC 6N, for those BGA money minting) so they worked with ASMedia to make the chipsets, and went with a 2x system rather than single Chipset, so there is a slight possibility of handoff issues, if they happen or other unknown variables.

 

So Ultimately the PCIe5.0 is downgraded to PCIe4.0 on Zen 4 launching boards regardless of X670E or the budget one B series. Because there's a massive cost there apparently as per the article author who has good comments there, as the Switch might cost in triple digits per unit. On top of the daisy chaining. So the Zen 4 chipset link is same as AM4 Zen 3 only.  What this means is Zen 4 is having same DMI speed like Zen 3 at 4 lanes 4.0 instead of 4 lanes 5.0 and then switch to 4.0 and split with PLX at Chipset (cannot do because high cost), unlike Intel which upgraded from CML 4 lanes 3.0 to RKL 8 lanes 3.0 to ADL 8 lanes 4.0.

 

Anyways thought something is interesting to learn, esp we know that HEDT is DEAD. X299 being last from Intel and half cooked Threadrippers. Rumors are next gen HEDT will be having W class from Intel (meaning no more X series and more Workstation BS) and Threadripper Pro meaning ultimate high price. So we need to know more about the Mainsteam Processors and their I/O capabilities.

[Mr. Fox]

25 minutes ago, Ashtrix said:

Anyways thought something is interesting to learn, esp we know that HEDT is DEAD. X299 being last from Intel and half cooked Threadrippers. Rumors are next gen HEDT will be having W class from Intel (meaning no more X series and more Workstation BS) and Threadripper Pro meaning ultimate high price. So we need to know more about the Mainsteam Processors and their I/O capabilities.

Tomorrow's best computers will most likely be built using today's best components. Computers are becoming more like cars. The newer they are, the crappier they are. It is rare to find examples of anything newer being better. And, by anything I literally mean anything (not just computers and cars). Newer almost always means compromised, less reliable, less durable, but more expensive.

[Papusan]

1 hour ago, Ashtrix said:

Both AMD and Intel are really massively deep into screwing customers. Shame that Intel is again following same garbage 2 CPU per socket formula as always while AMD is giving so many damn things. I think Intel might fix the ILM in a way it doesn't break the cooler compat with LGA1700 on upcoming Z790 if not they also should be avoided unless one knows how to fix the ILM and handle that DIY tweak.

What I don't understand... Why suddenly the need for 50% more pins on a minstream platform. Remeber AMD most liekly have to support AM5 several years down the road. Maybe to add in some gimmick that they need to offer to try compete with AMD's real cores? I just can't see why they need so many more pins on the socket. Or just because this is only a real scam to try lure people "why they need new MB"? 

 

More junk on the PCB may need those extra pins. 

https://wccftech.com/intel-14th-gen-meteor-lake-s-desktop-cpus-confirmed-in-leaked-drivers-tiled-chips-for-mainstream-pc-builders/

[Reciever]

4 hours ago, Ashtrix said:

 

Both AMD and Intel are really massively deep into screwing customers. Shame that Intel is again following same garbage 2 CPU per socket formula as always while AMD is giving so many damn things. I think Intel might fix the ILM in a way it doesn't break the cooler compat with LGA1700 on upcoming Z790 if not they also should be avoided unless one knows how to fix the ILM and handle that DIY tweak.

 

 

Intel is deadlocked into this 8P fixed and will keep on adding small improvements to the P cores while adding more E JUNK. Because their Core is same architecture. I think after 2026 they might do a full new micro architecture. Meanwhile Zen will bulldoze them, god I hope. I hate this E junk nonsense and giving 8P max for that gaming performance joke and milk more on BGA TRASH 🤬

 

Check this information out. I learned some information on AMD's upcoming Zen 4 Chipset. This lane business is always confusing no matter what lol...

 

AMD Zen 4 & Socket AM5 Explained: PCIe Lanes, Chipsets, Connectivity

 

 

 

 

 

Basically the X670E / X670 Dual Chipset Design has to prove it's Reliability - which AMD didn't show with AM4 as they are still busted with USB issues thanks to GloFo12nm IOD which is JUNK and doesn't work properly with Zen 3 chiplets, and Performance.

 

Also looks like AMD wanted to go somewhat cost effective route (why not when you spend tons of cash on the BS garbage dumpster class junk called BGA trash and give it TSMC 6N, for those BGA money minting) so they worked with ASMedia to make the chipsets, and went with a 2x system rather than single Chipset, so there is a slight possibility of handoff issues, if they happen or other unknown variables.

 

So Ultimately the PCIe5.0 is downgraded to PCIe4.0 on Zen 4 launching boards regardless of X670E or the budget one B series. Because there's a massive cost there apparently as per the article author who has good comments there, as the Switch might cost in triple digits per unit. On top of the daisy chaining. So the Zen 4 chipset link is same as AM4 Zen 3 only.  What this means is Zen 4 is having same DMI speed like Zen 3 at 4 lanes 4.0 instead of 4 lanes 5.0 and then switch to 4.0 and split with PLX at Chipset (cannot do because high cost), unlike Intel which upgraded from CML 4 lanes 3.0 to RKL 8 lanes 3.0 to ADL 8 lanes 4.0.

 

Anyways thought something is interesting to learn, esp we know that HEDT is DEAD. X299 being last from Intel and half cooked Threadrippers. Rumors are next gen HEDT will be having W class from Intel (meaning no more X series and more Workstation BS) and Threadripper Pro meaning ultimate high price. So we need to know more about the Mainsteam Processors and their I/O capabilities.

I wonder if its because i never updated my sBIOS version for why I never experienced the USB issues, Hiew either from what I can recall as our systems were similar, actually thats the same for most of our Teamspeak group now that I think about it (AM4, not same SKU)