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

[electrosoft]

Good news:

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

 

Been testing the TG vs Patriot sticks and the TG sticks are vastly superior.

 

Better XMP timings and lower voltage (1.4 vs 1.45) at the same speed.

 

TG passed 8000 90 min TM5 XMP right out of the box even hitting close to 70c no problem.

 

Patriot erred out at 8200, 8000 and on a long run of 7800 (before 90 min).

 

As a test, I set the Patriot sticks to TG timings and voltage and they erred out immediately at 8200, 8000 and within 10 minutes at 7800 with temps either only hitting 55c or in the case og 8200, 43c. They're garbage and going back.

 

Crazy how much beefier the TG heatsinks are. I weighed each stick and the Patriots weigh 47g/ea and the TG sticks are 84g/ea.

 

I'm most likely going to hold onto the TG sticks but.....

 

Bad news:

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

After more granular testing, I'm pretty sure this chip has the SA Bug. The TG sticks let me find it as the Patriot sticks would error out on TM5 before triggering it. TG sticks at 8000 and 8200 just sucked it up till the bug triggered (complete system lock up, black screen, requiring hard power off/on).

 

No problems with DDR4 at 1.35v and no problems at lower frequencies at 1.35SA which is must likely an issue of setting vs requested voltages. As I scale up, requested voltage triggers it and boom.

 

Once you're at 8000 and 8200 auto SA (or hard set SA) and you go to run TM5, instant (8200) or within 5 minutes (8000) lock up, black screen and power reset required.

 

Testing SA shows 1.18 is rock solid. 1.20 it will eventually SA Bug out. The higher you scale the quicker it triggers.

 

IMC on this chip is indicating 8200 @ 1.18v it will eventually hit a wall and need more SA and bug out. Hard set to 1.18v at 8200 and the system will just reboot 20-30 min into testing. I'll run a few more tests with a touch more Vmem just to completely rule out the sticks vs SA wall but either way, it looks like overall this chip is SA Bugged.

 

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

 

Not sure if I'm going to keep it but everything else about it seems pretty solid. I just have to decide if I'm good with the SA Bug as 8000 tuned was my D2D goal.

 

 

 

 

[Mr. Fox]

2 hours ago, electrosoft said:

Not sure if I'm going to keep it but everything else about it seems pretty solid. I just have to decide if I'm good with the SA Bug as 8000 tuned was my D2D goal.

I had one truly amazing Patriot Viper Blackout DDR4 Samsung B-die memory kit. Two others of the same SKU were trash and none of the Patriot DDR5 I have tested was any good. The TG Xtreme 48GB DDR5-8200 kit I have is excellent. Definitely superior to the the G.Skill. But, the TG Delta kits seem to get beaten by Trident Z kits.

 

The CPU I have with the SA bug is rock solid. I would buy it again if given the opportunity. No regrets. And, with tighter timings 8200 performs the same or better than 8400. So close there is no way to quantify a difference other than an AIDA64 memory benchmark and the difference is so small that the added hassle of tuning and higher voltage simply isn't worth caring about. Monumental waste of effort due to the absence of a tangible benefit in any workload scenario other than an AIDA64 benchmrk. I am sure it could go higher with the memory overclocking with more VCCSA, but even with the SA bug it runs better than most that don't have the SA bug. That's why I am running that CPU in the SFF system. Without water cooling for the memory there isn't any thermal headroom, which makes the SA bug irrelevant in practical application. 

 

Keeping everything in perspective, if you are getting 8000 or 8200 with under 1.200V VCCSA and are capped because of the bug, but the VID and Vcore needs are superior that is a way better CPU that one that is a VID pig that can run 8400, 8600 or 8800. I'd much rather have a strong core overclocker that doesn't crater itself under molten silicon due to insane voltage requirements.

---

I sure have had a nice time doing nothing today. Not ready to go back to the grind tomorrow. That old X79 setup with the Titan GPU does plenty of things "good enough" and I've wheeled it into the living room and hooked it up to the TV. Played some Modern Warfare and Crysis 3 Remastered today on it from the comfort of my recliner. I would never know it is averaging only 75 FPS instead of 175 FPS but for the OSD telling me so. For an "almost free" computer it's pretty awesome. I say "almost free" because I had to pay the postage and trade spare parts that I never had any intention of using. Close enough to free.

 

The only reason I get to do this is Mrs. Fox is away visiting her family. She would pitch a fit about me doing this if she were at home.

Crysis 3 Remasered - GtX Titan 1080p with medium settings...

 

[Mr. Fox]

I forgot to share this. Others may find it useful.

 

The Rampage IV Gene BIOS fan controls are poor (very limited) compared to newer ASUS motherboards. I found this application for manually controlling the AIO pump and fans on the Gene. It works fantastic. And, it is inexpensive. You can try the free version before buying. The default configuration is annoying because it has system tray icons (like Core Temp does) showing values for monitored sensors After fussing with the settings to turn them off and configuring he option to minimize to tray I am very pleased with it. I suspect it will work for most systems, including laptops.

 

I turned off everything except the fan controls. I don't need the monitoring for everything because I use HWiNFO64 for that.

 

https://www.argusmonitor.com/

 

[Tenoroon]

On 4/25/2023 at 11:24 PM, johnksss said:

 

I would need to check, but i may still have a heatsink cut out sitting in my heat sink bin.

I know this reply was super old, but it was in reference to @electrosoftmentioning you had modified a P870TM vapor chamber to fit in a Turing GPU. I purchased a 2070S MXM for 50 bucks and am trying to find how to modify the heatsink. I've been looking in the archive, but I can't find much regarding the modification.

If you have some time to spare, a picture would be greatly appreciated, if not or you no longer have the heatsink, all is well.

[electrosoft]

6 hours ago, Mr. Fox said:

I had one truly amazing Patriot Viper Blackout DDR4 Samsung B-die memory kit. Two others of the same SKU were trash and none of the Patriot DDR5 I have tested was any good. The TG Xtreme 48GB DDR5-8200 kit I have is excellent. Definitely superior to the the G.Skill. But, the TG Delta kits seem to get beaten by Trident Z kits.

 

The CPU I have with the SA bug is rock solid. And, with tighter timings 8200 performs the same or better than 8400. So close there is no way to quantify a difference other than an AIDA64 memory benchmark and the difference is so small that the added hassle of tuning and higher voltage simply isn't worth the effort. Monumental waste of effort due to the absence of a tangible benefit. I am sure it could go higher with the memory overclocking with more VCCSA, but even with the SA bug it runs better than most that don't have the SA bug. That's why I am running that CPU in the SFF system. Without water cooling for the memory there isn't any thermal headroom, which makes the SA bug irrelevant in practical application.

 

Yeah that is why I ordered the TG. At first I requested a replacement set of Patriots, but I decided to pony up a little more $$$ to truly test a different set and I'm glad I did.  Monitoring (VCC)SA @ 1.18 fixed = 1.161 actual during TM5 so I'm rerunning the 8200 tests at 1.42v vs 1.40v stock and logging the results to see if there is any spike/fluctuation.

 

Passed TM5 90 min @ 8200 XMP pushing to 1.45v

 

 

Booted and posted with 8400 and 8600 even with XMP enabled. I'll try seeing what 8400 can do over the next day or so and see how high it can boot/post. Patriot sticks wouldn't even post 8400 at 1.45v and loose 5600 Jedec timings....garbage sticks.

 

---

 

As for the CPU, that is what I'm kind of thinking too. How far will I be able to truly press my sticks in an SFF? I'd most likely end up having to scale back due to heat in a 12.4L case. I'm going to see how it handles 5.8 all core on an AIO. Based on 5.6 and 5.7 coasting along nicely, might be doable. Everything right now will continue to be tested at 8000 min and really dial that in tight. Dunno where I'll land a SP109 for what I paid and it can do 8200 and maybe more.

[tps3443]

17 hours ago, electrosoft said:

 

Yeah that is why I ordered the TG. At first I requested a replacement set of Patriots, but I decided to pony up a little more $$$ to truly test a different set and I'm glad I did.  Monitoring (VCC)SA @ 1.18 fixed = 1.161 actual during TM5 so I'm rerunning the 8200 tests at 1.42v vs 1.40v stock and logging the results to see if there is any spike/fluctuation.

 

Passed TM5 90 min @ 8200 XMP pushing to 1.45v

 

 

Booted and posted with 8400 and 8600 even with XMP enabled. I'll try seeing what 8400 can do over the next day or so and see how high it can boot/post. Patriot sticks wouldn't even post 8400 at 1.45v and loose 5600 Jedec timings....garbage sticks.

 

---

 

As for the CPU, that is what I'm kind of thinking too. How far will I be able to truly press my sticks in an SFF? I'd most likely end up having to scale back due to heat in a 12.4L case. I'm going to see how it handles 5.8 all core on an AIO. Based on 5.6 and 5.7 coasting along nicely, might be doable. Everything right now will continue to be tested at 8000 min and really dial that in tight. Dunno where I'll land a SP109 for what I paid and it can do 8200 and maybe more.

Give it some solid testing in games, make sure it doesn’t freeze with DDR5 8200 using your current SA voltage and rock on! (give it a few days of solid testing in lots of things) My other SP99 14900KS had an SA bug. That was mainly why I returned it though. It couldn’t daily 8400 so I let it go back, I didn’t mind the SP99 at all 🤣. Real world daily limited me to about DDR5 8200 range for true stability with no freezing on that chip too. You’d think it was okay and stable at 8400, and then Randomly one day, your PC locks up, and then again randomly a week later. I got lucky on this 2nd SP108 with no sa bug, but since your chip is sp109 and very good cores, and if it can do DDR5 8000+ I’d say that is absolutely acceptable for SFF build. You may even be able to run some crazy tight 8000c32 or 8200c34. 😍 

 

I’m all about sacrifices! And accepting weaknesses in a chip. They all have some sort of weakness anyways. 

[Papusan]

See from 1:11:52 - Badly done binning (make defective i9 instead of bin the chips down to i7), elevated voltage for every SKU and oxidation problem. 3 major flaws in a row. Yup, Intel have made it hard for themself. 

 

[electrosoft]

4 hours ago, Papusan said:

See from 1:11:52 - Badly done binning (make defective i9 instead of bin the chips down to i7), elevated voltage for every SKU and oxidation problem. 3 major flaws in a row. Yup, Intel have made it hard for themself. 

 

 

I believe Intel would coin this, "Aggressive binning" 🤣

 

How some of those chips made it out to a market dominated by air and aio cooling and joe average user that has no idea what they're doing is beyond me.....

[electrosoft]

Enforcing Intel's limits while pushing LLC to 5 (highest for Asrock to achieve minimum boost) and Vcore compensation to 1 (lowest assist). This is with per core auto engaged. It scores just about the same as Intel's defaults out the box and doesn't turn it into a smoldering firepit.

 

Vcore under load all core = 1.088 for Multi and 1.392 for single (<---this is Intel's problem on poorly binned CPUs  and/or just with basic enforements in place and those lower core runs shooting up to 1.5+). Single core was sitting at ~6.2ghz.

 

72c for multi, like 52c for single. Auto fans.

 

I can see why this Lian-Li Galahad Performance tops the charts routinely for AIOs. It's a monster.

 

[Papusan]

11 hours ago, electrosoft said:

 

I believe Intel would coin this, "Aggressive binning" 🤣

 

How some of those chips made it out to a market dominated by air and aio cooling and joe average user that has no idea what they're doing is beyond me.....

 

I wonder how it would go if you could rent processors from Intel. Maybe you could get their best binned Cpu's ? Return it to sender if it's trash (has to be within their best 2% best binned chips or back it goes). 

 

The tech trend continue rolling. Anyone want Subscription-based hardware?

 

 

 It's difficult seeing many computer users buying into this monetization model. 

 

Faber is more confident about this being the future of mice after visiting the company’s innovation center in Ireland. There, she says one of the team members showed her a concept of the “forever mouse” and described it as “a little heavier, it had great software and services that you’d constantly update, and it was beautiful.” The mouse is compared with a nice watch that isn’t just thrown away, such as a Rolex.

 

Logitech Floats Idea Of A Subscription-Based Forever Mouse With DLC

 

 

[Mr. Fox]

2 hours ago, Papusan said:

 

I wonder how it would go if you could rent processors from Intel. Maybe you could get their best binned Cpu's ? Return it to sender if it's trash (has to be within their best 2% best binned chips or back it goes). 

 

The tech trend continue rolling. Anyone want Subscription-based hardware?

 

 

 It's difficult seeing many computer users buying into this monetization model. 

 

Faber is more confident about this being the future of mice after visiting the company’s innovation center in Ireland. There, she says one of the team members showed her a concept of the “forever mouse” and described it as “a little heavier, it had great software and services that you’d constantly update, and it was beautiful.” The mouse is compared with a nice watch that isn’t just thrown away, such as a Rolex.

 

Logitech Floats Idea Of A Subscription-Based Forever Mouse With DLC

 

 

Sad that they would offer it, and sad that some people will actually go along with the program. We live in an era of rampant stupidity where intelligence is rare and idiocy and foolishness are becoming the new normal. 

[Papusan]

12 hours ago, Mr. Fox said:

Sad that they would offer it, and sad that some people will actually go along with the program. We live in an era of rampant stupidity where intelligence is rare and idiocy and foolishness are becoming the new normal. 

 

The latest from the techno industry: Mouse rental. Yep, this is disgusting. And worse it will be. Maybe because of stupid consumers that accept it. They will accept everything thats new/more modern. And the chiefs in tech companies know their audience. They know very well that a big part of their consumer base is damn stupid. 

 

The time BMW leased seat heating

This has occasionally led to a reputational crash when more physical products have been given a monthly price. For example, BMW was ridiculed by a near united internet when in some markets they set a monthly price for heated seats .

 

 

 

 

https://www.theverge.com/24206847/logitech-ceo-hanneke-faber-mouse-keyboard-gaming-decoder-podcast-interview

 

Hanneke Faber, you are the new-ish CEO of Logitech. Welcome to Decoder.

 

Thank you so much.

I am really excited to talk to you. You have been on the job as CEO for seven months now?

 

Nice....... A newish CEO try stand out🤢

 

What's next? If you buy an new ssd, then you'll have to pay a subscription to be allowed to store your data on it? Hmmm. 

 

My good old logitech mouse and KB will be the last one I get from Logitech if they go the subscription route forwards. Then goodbye. Now it's up to the rest of consumers. Because I don't wan't be a part of this crazy madness. 

 

Second bad idea.......

 

Newegg launches CPU trade-in program with low payouts — $300 for a Core i9-14900K or $220 for a Ryzen 7 7800X3D

 

The perfect timing. Right before AMD launch their next gen Ryzen and with Intel flooded with problems. But the trade-in program is an very bad idea whatever you go with Intel or AMD. Equal tasteless as Neweggs GPU trading program. 

 

If you're sold on Newegg's trade-in program and want even less money back for your tech, you could trade in an entire PC for a maximum of $400. Until August 11th, Intel is sponsoring a PC Trade-In deal allowing you to trade in a desktop, laptop, or tablet for up to $400 off the purchase of a new 14th-gen computer.

 

 

[cylix]

Significant lower price for the new Amd chips.

https://www.tomshardware.com/pc-components/cpus/amd-ryzen-9000-price-listings-now-on-best-buy

 

The prices range from $279 for the Ryzen 5 9600X to $599 for the Ryzen 9 9950X. Although the price tags look reasonable, we recommend approaching them cautiously since they could be placeholders.

 

Let's see how the new x870 mainboards will pe priced and also the x3d chips later.

[Papusan]

Also AMD failed with binning/QC problems. But will AMD tell their customers why they asked for a recall of all Zen5 chips? Probably not. Yup, if it wasn't for all Intel problems we would see trash binned Ryzen flagship out there now. So as I said before.... Saved by Intel😁

 

Intel should have done the same. Only top binned chips should be binned for i9's. And only the best 3% chips should be binned as KS. If they can't meet demand then honestly say it (consumers will understand) and don't do stupid tihings as binning i7's as i9's to increase profits.

 

 

 

 

From bad to worse... Maybe it's time to get rid of the CEO and baby-cores? Intel reminds me more about a sinking ship. 

 

Intel specifically called out looming cuts to marketing and R&D, the latter being a concerning development for a company that's reliant on developing new tech, along with reductions in general and administrative staff.

Intel to layoff more than 15% of workforce — almost 20,000 employees — encountered Meteor Lake yield issues, suspends dividend

 

Intel class action lawsuit investigation begins for the company's CPU crashing and instability issues

 

On 7/31/2024 at 1:21 PM, Papusan said:

I wonder how it would go if you could rent processors from Intel. Maybe you could get their best binned Cpu's ? Return it to sender if it's trash (has to be within their best 2% best binned chips or back it goes). 

 

The tech trend continue rolling. Anyone want Subscription-based hardware?

On 7/31/2024 at 4:31 PM, Papusan said:

The latest from the techno industry: Mouse rental. Yep, this is disgusting. And worse it will be. Maybe because of stupid consumers that accept it. They will accept everything thats new/more modern. And the chiefs in tech companies know their audience. 

 

More of the same. The trend continue...... If you don't have the money. No problem... Just rent what can't afford.  

NZXT Introduces Flex PC, Subscription-based Gaming PC Rent Program with Upgrades

Press Release  Today, 14:35

 

NZXT, a leader in PC gaming components, products and prebuilts, builds on its commitment to making PC gaming achievable to all of those with a passion for play with the introduction of NZXT Flex. A first of its kind gaming PC subscription service, Flex enables gamers to enjoy the benefits of a high performance gaming PC for a low monthly fee. The Flex subscription service features no hassle returns, no hidden fees and regular upgrades all without a long term commitment.

 

On 7/31/2024 at 4:31 PM, Papusan said:

https://www.theverge.com/24206847/logitech-ceo-hanneke-faber-mouse-keyboard-gaming-decoder-podcast-interview

 

Hanneke Faber, you are the new-ish CEO of Logitech. Welcome. Nice... A newish CEO try stand out🤢

My good old logitech mouse and KB will be the last one I get from Logitech if they go the subscription route forwards. Then goodbye. Now it's up to the rest of consumers. Because I don't wan't be a part of this crazy madness. 

 

Hey Logitech, every mouse should be a forever mouse

 

A product that you pay for should be supported, period, without the need for an ongoing payment.

 

All Logitech’s talk does is annoy people who have no desire for another subscription in the first place. And for anyone else considering the same strategy: I’d rather buy from your competition than deal with paying yet another monthly service charge. I suspect I’m not alone.

 

 

[cylix]

Bad times at intel indeed, that workforce cut is massive..really bad decision making across the board. Losing on all fronts. No worries @Papusan that Ceo will get a good chunk of money  if he goes away. He doesn't care. But the employees  and we the customers will get screwed again😑

[Papusan]

10 hours ago, cylix said:

Bad times at intel indeed, that workforce cut is massive..really bad decision making across the board. Losing on all fronts. No worries @Papusan that Ceo will get a good chunk of money  if he goes away. He doesn't care. But the employees  and we the customers will get screwed again😑

 

Here's somewhat good news. You all will get 5 years warranty (extended from 3 to 5 years or 3 years for the KS) from Intel for your boxed 13th and 14th gen processors (I only buy boxed Cpu's here home). Intel is forced to do so. Or they risk a lot more than what they experience right now. So slightly good news for you all that prefer running the 2 cores boost at sky-high voltage. This will also help if you later will sell your Cpu in the used market.

 

Intel announces an extra two years of warranty for its chips amid crashing and instability issues — longer warranty applies to 13th- and 14th-Gen Core processors

 

Intel trying to build goodwill.

 

Intel Via Oxidation Statement

The issue was identified in late 2022, and with the manufacturing improvements and additional screens implemented Intel was able to confirm full removal of impacted processors in our supply chain by early 2024. However, on-shelf inventory may have persisted into early 2024 as a result. Why in Hell not re call them right after their findings? They know very what batches was affected. Why use over 1.5 year to clean up their supply chain? They tried hide it under the bush because this would certainly hurt their reputation. Such move will always hit back. And they pay for it now. 

 

Additional info from Anandtech.

One important thing to note, however, is that the extended warranty will only apply to boxed processors, i.e. Intel’s official retail chips. Intel’s loose chips that are sold by the tray to OEMs and certain distributors – commonly referred to as “tray” processors – are not covered by the extended warranty. While Raptor Lake tray processors do technically come with a three-year warranty of their own, Intel does not provide direct, end-user warranty service for these chips. Instead, those warranties are serviced by the OEM or distributor that sold the chip.

With the bulk of Intel’s chips going to OEMs and other professional system builders, Intel will undoubtedly need to settle things with those groups, as well. But with OEM dealings typically remaining behind closed doors, it’s unlikely we’ll hear about just what is agreed there.

 

 

Loads of new complaints everyday on the internet forwards will undoubtedly hurt some of Intel's reputation. This RMA mess will pop up on all sorts of forum and tech press the next 2-3 years. Hope Intel soon will see the paint on the wall. Not all publicity about Intel's products is good publicity weeks after weeks forwards.

 

 

The RMA drama begins.... Tomshardware.com - Today

 

This sweet swan dive will be small compared to what to come, if they don't act correctly. 20.000 fired Intel employees can fast be the double....

 

To this effect, the company is launching its "Lunar Lake" and "Arrow Lake" processors within 2024, to address the various PC sub-segments. The Intel stock isn't churning in a silo, tech stock prices across the industry are witnessing corrections, although few as remarkable as Intel.

[tps3443]

5 hours ago, Papusan said:

 

Here's somewhat good news. You all will get 5 years warranty (extended from 3 to 5 years or 3 years for the KS) from Intel for your boxed 13th and 14th gen processors (I only buy boxed Cpu's here home). Intel is forced to do so. Or they risk a lot more than what they experience right now. So slightly good news for you all that prefer running the 2 cores boost at sky-high voltage. 

 

Intel announces an extra two years of warranty for its chips amid crashing and instability issues — longer warranty applies to 13th- and 14th-Gen Core processors

 

Trying to build goodwill.

 

Intel Via Oxidation Statement

The issue was identified in late 2022, and with the manufacturing improvements and additional screens implemented Intel was able to confirm full removal of impacted processors in our supply chain by early 2024. However, on-shelf inventory may have persisted into early 2024 as a result. Why in Hell not re call them right after their findings? They know very what batches was affected. Why use over 1.5 year to clean up their supply chain? They tried hide it under the bush because this would certainly hurt their reputation. Such move will always hit back. And they pay for it now. 

The 5 year warranty is nice, but dang that’s a long time to keep a cpu I think 2 years is stretching it 🤣 I also use the 6.2Ghz boost, but I always aim for running the lower load voltages and lower max amperage and lowest power. I checked my Vmin yesterday after 560+ hours on my SP108 KS. I set a straight 5.9P/4.5E auto ring (default clocks), and my Vmin was 1.190V in Cinebench R23 load (CPU is not delidded) From my experience the degradation range is 1.300V+ load voltages. Not from using the 1.400-1.500V+ idle VID’s with low amperage on 6.2Ghz (2) core loads.  
 

I think the biggest problem with a 14900KS is if I reset my bios and run an average chip its load R23 default voltage is beyond 1.320-1.350v + which is red zone for long term or short term, sub 1.300+ is enough to feed 6.1Ghz all core with good cooling, so with these 14900KS fighting for 5.9Ghz sustained on AIO’s they were climbing beyond 1.350-1.375+ load which is melting them down. (Newer bios have fixed this of course) that is the main high voltage issue. High Vcore under extreme loads or moderate loads long term. And high amperage under high heat. 

[electrosoft]

4 hours ago, tps3443 said:

The 5 year warranty is nice, but dang that’s a long time to keep a cpu I think 2 years is stretching it 🤣 I also use the 6.2Ghz boost, but I always aim for running the lower load voltages and lower max amperage and lowest power. I checked my Vmin yesterday after 560+ hours on my SP108 KS. I set a straight 5.9P/4.5E auto ring (default clocks), and my Vmin was 1.190V in Cinebench R23 load (CPU is not delidded) From my experience the degradation range is 1.300V+ load voltages. Not from using the 1.400-1.500V+ idle VID’s with low amperage on 6.2Ghz (2) core loads.  
 

I think the biggest problem with a 14900KS is if I reset my bios and run an average chip its load R23 default voltage is beyond 1.320-1.350v + which is red zone for long term or short term, sub 1.300+ is enough to feed 6.1Ghz all core with good cooling, so with these 14900KS fighting for 5.9Ghz sustained on AIO’s they were climbing beyond 1.350-1.375+ load which is melting them down. (Newer bios have fixed this of course) that is the main high voltage issue. High Vcore under extreme loads or moderate loads long term. And high amperage under high heat. 

 

Cooling really helps as does silicon quality. On an AIO, I am running 5.6/4.5E auto ring and 5.7/4.5E auto ring. I'm dialing in a 5.8 config of the same nature too. Bios set = 1.28/1.33. Looks like 5.8 will dial in around 1.38v BIOS but not finished testing yet.

 

5.6 all core load = 1.152v on an AIO. Max temp = ~70

5.7 all core load = ~1.200v on an AIO. Max temp = ~76

5.8 all core load  = 1.232v on an AIO. Max temp = ~81

 

This was all using Falk's testing parameters from the second post over on the OCN forums

 

I only test with fans on auto never max. While this may skew the initial temps with a spike before ramp, it's more realistic for my use case.

 

I'm tempted to see what a 5.9 all core load will look like on this SP109 on an AIO since 5.6->5.8 is doable. This is with basic CB23 testing.

 

What's your V/F curve look like for your SP108? Post a pic.

 

---

 

I don't think 1.35v-1.4v is the problem. I think a massive series of transient spikes and/or 1.5+ temporary loads over and over again are the problem.

 

I expect Intel to cap the VR out max to ~1.45v with the MC update. *Maybe* 1.40v at the worst for over compensation. As it stands, we are seeing single core boosting to 1.5+, 1.6 and sometimes a bit higher which is insane and the real problem IMHO. You start to damage your preferred cores and all hell breaks loose. You force multiple cores to potentially pull those insane volts and it's game over sooner than later. With this limitation, there are suddenly going to be a rash of chips that can't hit their 6.2 targets consistently.

 

This would also show reason why higher binned silicon isn't degrading as fast or actually surviving because they don't need to boost that high unless you get those who are stuck on wanting to run 6.0+ all the time single and multi 24/7.

[cylix]

How slimy can Intel get????🤮

 

Intel has denied two of my 14900K RMAs (instability) and stated they will confiscate or destroy them if I proceed with the warranty process.

 

https://www.reddit.com/r/LinusTechTips/s/dXBEzcXPdA

[ryan]

Better than Lenovo. I returned two tablets 2 months ago and still haven't got a refund, they said they refunded them and gave me a bogus receipt. Opened a dispute with bank, got a temp credit but unsure if they'll claw it back...expensive mistakes going with lenovo

[Mr. Fox]

3 hours ago, electrosoft said:

I don't think 1.35v-1.4v is the problem. I think a massive series of transient spikes and/or 1.5+ temporary loads over and over again are the problem.

 

I expect Intel to cap the VR out max to ~1.45v with the MC update. *Maybe* 1.40v at the worst for over compensation. As it stands, we are seeing single core boosting to 1.5+, 1.6 and sometimes a bit higher which is insane and the real problem IMHO. You start to damage your preferred cores and all hell breaks loose. You force multiple cores to potentially pull those insane volts and it's game over sooner than later. With this limitation, there are suddenly going to be a rash of chips that can't hit their 6.2 targets consistently.

 

This would also show reason why higher binned silicon isn't degrading as fast or actually surviving because they don't need to boost that high unless you get those who are stuck on wanting to run 6.0+ all the time single and multi 24/7.

Yes, I agree.

[electrosoft]

SP109 14900KS + TG 8200 + Asrock Z790i update:

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

 

So I've been testing UE5 shader compiles with Fortnite (clear cache, relaunch, repeat) with the 14900KS and whew, watching this suck down 300w+ and all cores ablaze on launch for a bit is something I've never seen in a game before but I never played some of the major heavy hitters like this before either. I don't remember CP2077 doing this (but that was on my 12900k). My 7900XTX will not fit in my test case since the 360mm AIO is mounted to the front so I slapped in my 3060ti from my backup 12400 system for testing. When I switch over to the Nucleus 240mm for phase 2 testing, I'll slap in the 7900XTX and swap back and forth with the 4090.

 

I think I just need to pick up an open bench and keep it for testing hardware. With modern hardware, this Corsair 540 is showing its volume age.

 

I haven't really used the Epic Game store and it is very unpolished with issues compared to Steam. Not a fan.

 

As for the Asrock Z790i Lightning, I like it, but I do have some pet peeves with it and little bugs that are annoying.

---

#1. The CMOS battery is glued to the side of the rear I/O with a 2 position connector wire that plugs in that is just screaming it will break eventually as everything you have to yank the wire to do a hard reset.

 

#2.  There is no IA VROUT MAX control like on Asus and Gigabyte.

 

#3. Ring will cap at 4.5ghz when selecting all core no matter what. You have to go in and select per core and set your all core that way for it to boost to 5ghz

 

#4. Whenever you change your fixed Vcore, it will automatically switch back to LLC1

---

 

 

Right now, I have 5 profiles setup for testing/use

 

56x 45x Auto Ring 8200 Fixed 1.28

57x 45x Auto Ring 8200 Fixed 1.33

58x 45x Auto Ring 8200 Fixed 1.37

59x 45x Auto Ring 8200 (work in progress) But right now looking like 1.41 atm

Auto freq w/ Enforced Intel limits with adjusted AC_LL/DC_LL and LL/Vcomp set to min 8200

 

Until the Intel MC update, I'm sticking with my fixed Vcore profiles. After we find out their official cap, I'll start dialing in the Auto freq profile to fully test 6.2 boosting and various games for data collection. I know I boost only to 1.392v max 6.2 auto CB23 but it isn't worth taking a chance just to be safe till then also knowing I prefer fixed overall anyhow.

 

All are working on AIO for my use cases.

 

All do not thermal throttle and range anywhere from ~68-70c (5.6x fixed along with Enforced) up to ~87c (5.9 fixed) with CB23 loop testing.

 

I still need to test 8400. Trying to boot 8800 = hard lock up requiring a physical power reset which means it is the SA Bug kicking in. 8600 boots but as soon as I run TM5, SA Bug lock up. This shows the limitations (if you can call them that for my meager setup) but also shows really good potential for the TG sticks and this Asrock Lightning Z790i board. I'd love to test a known, good chip up to 8800+ just to see.

 

Oh, and since I was testing shader compiles in Fortnight, I decided to try and play (I've never played before) and my nephew (who is a die hard player along with COD) laughed so hard at me failing over and over again in Battle Royale as I was getting obliterated left and right while trying to convince him back in the day I was an upper level Quake player who routinely pwned...
🙂

 

I mean just destroyed.....

 

😁

 

[Talon]

 

 

@Mr. Fox made the video. Are you going to request some royalties? 

[Mr. Fox]

3 hours ago, Talon said:

@Mr. Fox made the video. Are you going to request some royalties? 

I saw my posts and forum name and avatar several times at both EVGA and OC.net forums. 🤣

 

No matter what degree of merit any of this new round of psychotic pablum has, all of the negative publicity is going to be harmful to Intel and it could ultimately be more harmful to overclocking enthusiasts if they start locking things down and interfering with user tuning controls. The media is a two-edged sword and they have the power to create more issues than they solve. GN is certainly a trusted source and I do not doubt the accuracy (very well supported by facts) but I do question whether or not any of it is going to benefit anyone and may end up being more harmful than helpful in the long run.

 

One of my biggest take-aways from all of the hype (and it surfaced a few times in the GN video) is the hidden nugget (unintended and inconvenient truth) that refusing BIOS and OS updates when my systems are functioning the way I want them to is an extremely wise personal best practice that I have abided by for more than a decade and I shall continue doing so indefinitely because it has exempted me, almost entirely, from the stupid nonsense that update evangelists have suffered from. If it ain't broke, don't fix it. If you do, it will eventually get broken.

 

Nothing I have seen so far changes my fondness for my nicely binned 13th and 14th Gen i9 CPUs and none of it makes me pause to consider the possibility of AMD as a viable option. All of these big tech companies are equally dishonest scumbags that are rotten from head to toe and never let an opportunity to cover their own asses with lies and misrepresentations to slip by them. None of them give a rat's butt about their employees or the people that buy their products. We cannot trust any of them to be honest about anything... ever. The current mess with Intel causes too many of us to forget the past messes, missteps lies and scams from AMD and NVIDIA. Some fairly recent.

---

Speaking of lies and missrepresentations, for those who have drank the Kool-Aid and joined the muppet show of Winduhz 11 apostles, here is some fuel for the fire for those of us that despise the new cancer OS. Windows 11 offers no benefit and is actually harmful to performance. The only thing Winduhz 11 offers is a new level of ugliness for those that prefer a less-useful and tackier-looking GUI. The joke is on Windoze 11 users, but it's not a funny one. Utterly worthless OS downgrade. But, everyone is too busy hating on Intel to talk about this right now.

 

[electrosoft]

Monstrous value here used but like new sold and shipped from Asrock 6950XT for $420 via Amazon

 

Glad to see Asrock and Amazon worked out their differences as their store is nearly fully stocked with their products.
🙂