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

[electrosoft]

37 minutes ago, Raiderman said:

@electrosoft @jaybee83 45% is a whopping statement! I'll believe it when I buy it!

https://wccftech.com/amd-ryzen-9-9950x-16-core-zen-5-cpu-aida64-benchmarks-ddr5-8000-45-percent-faster-7950x/

 

I'll be real curious to see the entire breadth of benchmarks for the 9950x. We know it won't be as fast as the 7800X3D for gaming but in other memory intensive tasks, I'm real curious to see how much improvements they've made with the IMC to take full advantage of that memory bandwidth meaningfully whereas the 7000 series can now hit 8000+, but outside of memory bandwidth benchmarks, we're just not seeing any real tangible gains and in some cases performance loss between G2 and G4 (which makes sense when you  think at it).

 

I'm definitely excited to see what it brings to the table and if a 9950X3D is in my future. 🙂

 

I'm equal parts interested to see how 7000 and 9000 perform on 600 and 800 series boards too and if there is any performance loss using a 9000 series on a 600 series board.

 

 

[Raiderman]

16 minutes ago, electrosoft said:

 

I'll be real curious to see the entire breadth of benchmarks for the 9950x. We know it won't be as fast as the 7800X3D for gaming but in other memory intensive tasks, I'm real curious to see how much improvements they've made with the IMC to take full advantage of that memory bandwidth meaningfully whereas the 7000 series can now hit 8000+, but outside of memory bandwidth benchmarks, we're just not seeing any real tangible gains and in some cases performance loss between G2 and G4 (which makes sense when you  think at it).

 

I'm definitely excited to see what it brings to the table and if a 9950X3D is in my future. 🙂

 

I'm equal parts interested to see how 7000 and 9000 perform on 600 and 800 series boards too and if there is any performance loss using a 9000 series on a 600 series board.

 

 

I agree on the motherboard aspect. I'm hoping there isn't too much difference between the 600, and 800 series, because I finally have many of the bugs ironed out with my X670. It's been quirky, to say the least, and as much as I'd love to have a shiny new taichi x870, I'd prefer to hang with this gen for maybe another year.

[Mr. Fox]

The statement "faster in gaming" being tossed around like it is some kind of magical thing makes me kind of sad. It is as if the whole world has lost sight of what is most important. But, alas, they have. In countless ways. But, limiting conversation to computers rather than generalizations about life in 2024, the measure of a CPU that is most important should be how it does at things that only a CPU can do, not have it behaves when used to play games. It feels like a cop out or a distraction to be focused on how a GPU-centric task like gaming responds to a CPU when "better" isn't anything genuinely earth-shattering.  If the 9950X does better at workloads only a CPU can do it will be the better CPU than a 7800X3D even if the 7800X3D gets a few more FPS.

[Raiderman]

4 minutes ago, Mr. Fox said:

The statement "faster in gaming" being tossed around like it is some kind of magical thing makes me kind of sad. It is as if the whole world has lost sight of what is most important. But, alas, they have. In countless ways. But, limiting conversation to computers rather than generalizations about life in 2024, the measure of a CPU that is most important should be how it does at things that only a CPU can do, not have it behaves when used to play games. It feels like a cop out or a distraction to be focused on how a GPU-centric task like gaming responds to a CPU when "better" isn't anything genuinely earth-shattering.  If the 9950X does better at workloads only a CPU can do it will be the better CPU than a 7800X3D even if the 7800X3D gets a few more FPS.

I agree wholeheartedly with that statement, and that's why I stick with the standard Ryzen. On that note, there's this also.

https://www.tomshardware.com/pc-components/cpus/amds-ryzen-9000-wont-beat-the-previous-gen-x3d-models-in-gaming-but-theyll-be-close-improved-3d-v-cache-coming-too

[Papusan]

8 hours ago, Mr. Fox said:

The statement "faster in gaming" being tossed around like it is some kind of magical thing makes me kind of sad. It is as if the whole world has lost sight of what is most important. But, alas, they have. In countless ways. But, limiting conversation to computers rather than generalizations about life in 2024, the measure of a CPU that is most important should be how it does at things that only a CPU can do, not have it behaves when used to play games. It feels like a cop out or a distraction to be focused on how a GPU-centric task like gaming responds to a CPU when "better" isn't anything genuinely earth-shattering.  If the 9950X does better at workloads only a CPU can do it will be the better CPU than a 7800X3D even if the 7800X3D gets a few more FPS.

People being mad at me when I say that the Ryzen X3D chips is gamer cpus. We will see if AMD will let them clock equal high as vanilla Ryzen chips. But I doubt it. 

 

CUDIMM Standard Set to Make Desktop Memory a Bit Smarter and a Lot More Robust

 

For the storage jockey @Rage Set Not so sure this is very tempting. Or is it?😀

Seagate opens an eBay store to sell refurbished hard drives

To expand its Hard Drive Circularity Program and reduce electronic waste, Seagate is now selling its refurbished and recertified mechanical hard drives through eBay for attractive prices.

 

 

[cylix]

9 hours ago, electrosoft said:

 

I'm equal parts interested to see how 7000 and 9000 perform on 600 and 800 series boards too and if there is any performance loss using a 9000 series on a 600 series board.

 

 

 

I will be switching for sure for a x870 board and new memory which i think well be EXPO 6400 at least. I am now on the b650 and a new gen x870e will be a good upgrade.  Until now i like the Gigabyte x870e Aourus Pro ICE, clean white MB which will pass nice for my white build. But lets see what MSI brings out.

[Mr. Fox]

I was finally able to get the little munchkin on bare die. This reduced the core max 8 to 10°C over the already delidded CPU with a lapped core and IHS on liquid metal. I could see this being easily an 18 to 20°C improvement over a stock soldered IHS. There aren't many SFF systems capable of what this one is. Heck, it's better than most full-size builds, LoL. I am really impressed with what EKWB did here.

 

[Talon]

23 minutes ago, Mr. Fox said:

I was finally able to get the little munchkin on bare die. This reduced the core max 8 to 10°C over the already delidded CPU with a lapped core and IHS on liquid metal. I could see this being easily an 18 to 20°C improvement over a stock soldered IHS. There aren't many SFF systems capable of what this one is. Heck, it's better than most full-size builds, LoL. I am really impressed with what EKWB did here.

 

 

Nice! I really wanted to snag one of those EK Direct Die kits. 

[Mr. Fox]

2 hours ago, Talon said:

 

Nice! I really wanted to snag one of those EK Direct Die kits. 

Thank you. I bought this used at overclock.net because the one I tried to order from EKWB kept getting pushed out and after waiting more than 2 months I got a refund and cancelled. This one does not have the original fans so I am probably going to move the three fans attached to the chassis to the AIO and move the AIO fans to the chassis. The Bykski fans that came with the AIO are not square, so there is air space around them to leak static pressure. I am not sure why the person I bought it from did not have the original fans. I like the Bykski fans otherwise and it is a shame they are not square so they can seal off against one another. This is yet another example of a popular but stupid trend that demonstrates the sheeple are focused on form over function. They have a nice remote to control lighting and fan speed manually, which is my preference over firmware or software controls. I do not use RGB headers on any of my motherboards because I don't want to have to install the cancer bloatware needed to use them. All lighting on my systems is managed with molex or SATA powered ARGB controllers with a remote control. It's pretty sweet adding manual fan speed control to that.

 

[Rage Set]

5 hours ago, Papusan said:

People being mad at me when I say that the Ryzen X3D chips is gamer cpus. We will see if AMD will let them clock equal high as vanilla Ryzen chips. But I doubt it. 

 

CUDIMM Standard Set to Make Desktop Memory a Bit Smarter and a Lot More Robust

 

For the storage jockey @Rage Set Not so sure this is very tempting. Or is it?😀

Seagate opens an eBay store to sell refurbished hard drives

To expand its Hard Drive Circularity Program and reduce electronic waste, Seagate is now selling its refurbished and recertified mechanical hard drives through eBay for attractive prices.

 

 

 

Brother, my storage arrays are limited by cooling, otherwise I'd have a SAN set up in my homelab. Strictly speaking of raw storage, I'm in the 500TB+ club. Not including any of my SSD storage arrays.

 

All of this requires a lot of power but that's where solar comes into play. Some of you don't know but I'm a huge Prepper.

 

Back into the lurking shadows I go for now.

 

 

[Mr. Fox]

1 hour ago, Rage Set said:

Some of you don't know but I'm a huge Prepper.

All of us should be, even those of us that are not. We don't know when they will arrive, but darker days than we have already experienced are coming. The signs are there to see for anyone paying attention. Another four years like the last catastrophic four might be all it takes to cause the house of cards to fall. Unprecedented levels of ethical failure, moral decay, corruption and incompetent government, in the US and the world abroad. 

[tps3443]

23 hours ago, Rage Set said:

 

I have a lot of experience with them in an enterprise environment and "proconsumer" homelab. The one drive I really pushed people to get was the Intel P4510 8TB when it was down to $399. While not as fast the 5520, it featured the same full 1DWPD as the 5520. 

 

As you stated in your bulletpoint list, these "older" drives are great and there is no downside besides price...well there is the lack of real warranty (Solidigm took over Intel's drive business, but you have to purchase from them to get the warranty). 

 

Like the P4510, people like myself are waiting for the 5520 to hit $500 or below to pounce. 

Yeah I’ve been doing a lot of research on these drives. I’ve decided I am going to get an Intel Optane. I’m on the hunt for an Intel P5800X to use as a OS drive only. I’m trying to find a good deal. If I can’t find one here soon, I will probably just settle and grab a Optane 905P which are only $300 on sale. I really want the gen4 Optane though. I want to see what that ultra low latency feels like as an OS drive. I was also looking at these Intel Optane P5801X these are like a M.2/gum stick version of the U.2 P5800X. I have seen those for much cheaper but I’d need a E1.S to Gen4 PCIx4 adapter card. Which I may go that route as well. All of this is purely for fun of trying something different, but also because I keep hearing people say how amazing these Optanes are as an OS drive because of that crazy low latency and very high iops . 

[tps3443]

8 hours ago, Mr. Fox said:

I was finally able to get the little munchkin on bare die. This reduced the core max 8 to 10°C over the already delidded CPU with a lapped core and IHS on liquid metal. I could see this being easily an 18 to 20°C improvement over a stock soldered IHS. There aren't many SFF systems capable of what this one is. Heck, it's better than most full-size builds, LoL. I am really impressed with what EKWB did here.

 

Man I really like this! I love the skull on it looks really cool. This would honestly be a great way to get in to direct die very affordably. Someone could chop in to it and splice in to this loop with more pumps and bigger or more rads. 😎

 

[electrosoft]

18 hours ago, Raiderman said:

I agree on the motherboard aspect. I'm hoping there isn't too much difference between the 600, and 800 series, because I finally have many of the bugs ironed out with my X670. It's been quirky, to say the least, and as much as I'd love to have a shiny new taichi x870, I'd prefer to hang with this gen for maybe another year.

 

Ditto. I have no desire to upgrade my MSI board and it is running just fine atm (knock on wood). I'll be holding off till X3D so I'll get to get a taste of launch results from you and @jaybee83 at least.

 

18 hours ago, Mr. Fox said:

The statement "faster in gaming" being tossed around like it is some kind of magical thing makes me kind of sad. It is as if the whole world has lost sight of what is most important. But, alas, they have. In countless ways. But, limiting conversation to computers rather than generalizations about life in 2024, the measure of a CPU that is most important should be how it does at things that only a CPU can do, not have it behaves when used to play games. It feels like a cop out or a distraction to be focused on how a GPU-centric task like gaming responds to a CPU when "better" isn't anything genuinely earth-shattering.  If the 9950X does better at workloads only a CPU can do it will be the better CPU than a 7800X3D even if the 7800X3D gets a few more FPS.

 

Task is irrelevant. What matters to you is what is important and if increased CPU performance is a metric of validity. You bench. It really has no real world, tangible use except its a leisurely pursuit of fun to push hardware as much as possible in your level of benching (above most, below  LN2'ers) but I think you said it yourself that if you ditched benching because of shifting hardware market priorities by companies, you would most likely just run a turdbook as it is close enough for your actual D2D needs.

 

Gaming (and a close second is compile times) is my priority in regards to computational power and if the CPU is a limiting factor / bottleneck, I will start to seek out upgrades to improve it. If I ditched gaming and casual overclocking, I would probably drastically downsize my computer hardware and effectively run on a turdbook too along with my Macbooks for work. 🙂

 

9 hours ago, cylix said:

 

I will be switching for sure for a x870 board and new memory which i think well be EXPO 6400 at least. I am now on the b650 and a new gen x870e will be a good upgrade.  Until now i like the Gigabyte x870e Aourus Pro ICE, clean white MB which will pass nice for my white build. But lets see what MSI brings out.

 

Coming from a B650, that will be a most righteous upgrade! Do you plan on upgrading to the 9000 series too?

 

8 hours ago, Mr. Fox said:

I was finally able to get the little munchkin on bare die. This reduced the core max 8 to 10°C over the already delidded CPU with a lapped core and IHS on liquid metal. I could see this being easily an 18 to 20°C improvement over a stock soldered IHS. There aren't many SFF systems capable of what this one is. Heck, it's better than most full-size builds, LoL. I am really impressed with what EKWB did here.

 

 

Agreed 100% and really liking the results on the DD CR360. So many were whining and going on about the hotbox / thermal nuclear status of 13th and 14th gen, but getting one tamed quite nicely in a 18.4L SFF on a 240mm CR240 and running full tilt no problem was rather satisfying and on that same MSI  motherboard no less. 🙂

 

Equally satisfying was taming the X170SM/KM w/ 10900k and 11900k and the NH55 w/ 12900k too. There's an equal satisfaction getting hardware to run well in small form factors as there is having full wide open spaces and hardware and pushing it without restrictions. At least to me there is.

 

I actually picked up one of these for more SFF (open box for $90) adventures going from 18.4L to 12.4L  in an even smaller and tighter space.

 

What is the volume size of that case? It kind of looks like the volume of an NR200MAX

 

 

[Mr. Fox]

47 minutes ago, electrosoft said:

What is the volume size of that case? It kind of looks like the volume of an NR200MAX

 

https://www.amazon.com/gp/product/B0C5LK9L62/ref=ppx_yo_dt_b_search_asin_title?ie=UTF8&th=1

It is 33L. 8.7"D x 18.1"W x 13.7"H. For a SFF case it is unusually compelling in terms of capabilities that are usually totally absent or severely compromised. This actually makes it tolerable.

Quote

• Optimized for Cooling: With support for 280 and 360mm radiators and up to six fans, the AP201 is primed to deal with the thermal onslaught of high-performance hardware.
• Industry-Leading Spatial Efficiency: Despite a 33L footprint, the AP201 supports ATX PSUs up to 180 mm long, graphics cards up to 338 mm long, custom liquid cooling, and various storage devices.

 

[Papusan]

Hmmm. EVGA bake their cards as warranty fix. A repair afterwards won't be an easy fix or even possible. Interesting. The lazy man’s work to cut costs. Disgusting.

 

 

[cylix]

Pff, this is sad. How the mighty has fallen... I wonder if we will find out why EVGA went so hard down. I heard that the owner didnt want to sell out but right now looks like just a derelict ship waiting to sink. No swan song for them.

 

@electrosoft yeah ill get the 9000 but waiting for some more infos on the x3d chips, there are some rumors that both ccds will be the same with cache lvls. That will be awesome if true.

[Mr. Fox]

LOL. I saw this posted on Discord. 🤣

[jaybee83]

shoot, had connection issues and my reply got lost....

 

anyways, long story short:

 

- X670 vs. X870 are basically identical, the latter only has higher-end minimum requirements set by AMD

- IO die was stated to be "overall the same" on Zen 5, but that leaves wiggle room for finetuning, i.e. higher real life benefits from higher DDR5 speeds

- im expecting improved RAM support to come from the CPUs, not the mobos (notwithstanding 2 DIMM / higher-quality trace boards, but that goes for both X670 and X870)

- ive since switched basically all my high-intensive CPU workloads to the GPU, so im more open minded to at least consider the X3D skus this time around.

 

ill take my time and check the reviews first before making a purchase. good thing is, once the X3D chips are released, the regular chips will likely fall in price, so no real loss 🙂 

 

(ill still pull the trigger on the 5090 on launch day tho 😄 wait and see game will be CPU only 😛)

[Papusan]

Hmmm. What a mess if this is true. Yup, AMD is known for bad QC, QA and QM (see my older replies on that topic). Why on earth fiddle with the TDP for an X Sku in first place? Hope prosumers want 65W target because it't easier to cool? Or upsell more expensive SKUs?

 

Based on the information we have learned, AMD has updated its partners with a possible spec change for the 8-core part. It's being said that the AMD Ryzen 7 9700X might get a TDP of 120W which is almost 2x its current and official figure of 65W. 

 

AMD Might Re-Spec Ryzen 7 9700X “Zen 5” CPU With Higher 120W TDP To Outpace Its 7800X3D Chip

 

And Asus strikes again. What's wrong with Asus and Jayz ? @Mr. Fox 

 

[tps3443]

@Mr. Fox I’m wondering if you could help me with this.

 

I wanted to ask a question. I cannot find this answer anywhere about burification on the Apex. If I had a X8 lane PCIe add in card and wanted to run (2) x4 NVME's on this add in card using my 2nd PCIe x16 slot on my Z790 Apex, would that work? I know it would drop my top GPU PCIe to X8, but I'm wondering if the 2nd PCIe would see both of the x4 NVME SSD's in my 2nd x16 PCIe slot. 
 

I can’t seem to find this information. I see people reporting that it may only see (1) of them. 
 

 

[Mr. Fox]

6 hours ago, tps3443 said:

@Mr. Fox I’m wondering if you could help me with this.

 

I wanted to ask a question. I cannot find this answer anywhere about burification on the Apex. If I had a X8 lane PCIe add in card and wanted to run (2) x4 NVME's on this add in card using my 2nd PCIe x16 slot on my Z790 Apex, would that work? I know it would drop my top GPU PCIe to X8, but I'm wondering if the 2nd PCIe would see both of the x4 NVME SSD's in my 2nd x16 PCIe slot. 
 

I can’t seem to find this information. I see people reporting that it may only see (1) of them. 
 

 

Yes, it should work. I have PCIe NVMe add-in cards in both of my Apex motherboards. The cards of both dual and quad SSD configuration. The bifurcation works automatically. There's nothing to manually adjust. The feature is not specific to individual PCIe slots.

For PCIe 4.0 you would need one of these. https://www.amazon.com/gp/product/B0C64FVVHY?th=1

 

I don't really care whether my NVMe SSDs run at PCIe 3.0 or 4.0 speeds. PCIe 3.0 is fast enough.

 

I have one of these in my Apex: https://www.amazon.com/gp/product/B0BRYQH443?th=1

 

And, two of these  in my Apex Encore: https://www.amazon.com/gp/product/B09NQGZ2XJ

 

  

10 hours ago, Papusan said:

And Asus strikes again. What's wrong with Asus and Jayz ? @Mr. Fox 

 

ASUS motherboards have always been unreliable and unpredictable compared to others, and the one AMD motherboard I owned was never reliable. I am not sure what you are asking about Jay. If he wants extreme cooling with dry ice he needs to use a DICE pot and put it on the CPU. Trying to use a conventional loop makes no sense.

[Papusan]

2 hours ago, Mr. Fox said:

ASUS motherboards have always been unreliable and unpredictable compared to others, and the one AMD motherboard I owned was never reliable. I am not sure what you are asking about Jay. If he wants extreme cooling with dry ice he needs to use a DICE pot and put it on the CPU. Trying to use a conventional loop makes no sense.

 

It seems Jayz have problem with many of his Asus boards😎 And yup, that long loop makes no sense. But I think that idea was only for entertainment. Maybe he expected a desent drop in temp. But as you said, it won't work that way. And he should have used another type coolant with less viscosity for his project if he wanted some success.

[Mr. Fox]

28 minutes ago, Papusan said:

But I think that idea was only for entertainment.

Yes, I am confident of that. While I enjoy his videos, most of them are more focused on entertainment. Same can be said of Linus. And, that's OK. If they were not entertaining most people would not watch them. Buildziod is the opposite. His videos are very educational and useful, but difficult to watch because they are long, rambling, dry and not particularly entertaining. But, I usually learn something from them. I can't remember ever learning something new from watching Jay's or Linus's videos. I probably have a few times, but I can't remember an example of where I did.

[Papusan]

On 5/7/2024 at 4:19 PM, Papusan said:

Todays tech looks more and more like a tragic drama.... 

 

This below is taken from the link  I posted to @Talon 

 

Those that bought the $50 cheaper versions won't get any love from MSI due their greedy and disgusting tactics. All hardware should come with an opportunity to upgrade firmware to fix own trashy engineering/cheapo solutions.

 

The change-log follows.

FW.010 Firmware update information

• DSC switch added to the OSD to enable DLDSR and DSR modes.
• MPG 271QRX QD-OLED: Added HDMI PC/Console switch in the OSD. (MPG 271QRX QD-OLED default setting is "Console" mode.)
• Support for a variety of aspect ratios, MPG 321URX QD-OLED allows selection between a 24.5" and 27" option, MPG 271QRX QD-OLED allows 24.5".
• Fixed Color Gamut clamping while HDR is enabled. HDR mode will now store separate present mode settings.
• Optimized multiple language issues.
• Fixed image flicker issue when VRR is enabled.
• Added "Power LED" switch in the OSD to allows users to turn it off.
• Input lag and latency optimization.
• Fixed problem with white line during 2560x1440 @ 360 Hz. (MPG 271QRX QD-OLED only).
• Fixed displayed abnormal (dark screen) after VRR switched.
• Optimize a behavior of panel protect while protect completed that depends as executing power states.

 

League of Legends receives controversial Vanguard anti-cheat, Windows 11 now requires TPM 2.0

Riot denies reports of bricked PCs and clarifies its anti-cheat strategy...

Intel reportedly demands all board partners implement Intel Default (Baseline) Profile by May 31 — company hopes to fix issues with "some" Core i9 chips

Hmmm. People seems to forget that Intel will offer much higher OC headroom now for their audience😆 From 5.1Ghz and up to 6.0Ghz is more like wopping 17% oc head room. Many years since we saw Intel offered this much. Also genious cripple the performance for the KS so next gen  looks prettier on paper. All this drama because they went idiotic in the binning game.

 

See yourself... "Some" Core i9 chips. That mean Intel had to destroy for the waste majority of gamer boys and girls that don't overclock their hardware. They often go after highest possible clock-speed witthout have to overclock. At least you get above 5.0GHz out of the box from Intel now. The 6.2GHZ boost will turn into 5.1Ghz tragedy if they will implement settings as rumored. Not a problem for oc'ers but those are in minority. The majority of gamer kids will get castrated processors.

 

Intel want to set up a basic profile with a PL1 at 125W, a PL2 at 188W and an Iccmax at 249A. Clearly, if this is done as suggested, Intel will slash the performance down the drain for their processors. This is worse than expected. Much worse. This means Intel know very well there is some trash binned Core i9 out there. And RMA and sinking reputation cost money.

 

 

 

Exactly.

 

MSI Releases MPG 271QRX and MPG 321URX QD-OLED Firmware Update, Gives DSC Control

 

MSI forced by the community? Or hope that the screwed one will buy new hardware from MSI when the old need to be replaced? You can't trust MSI more than your trust on Asus or Microsoft. They all (tech companies) have their own agenda. They will do whatever serve them best. Can't be trusted  @Talon

 

MSI changes course: MAG QD-OLED monitors will get firmware updates after all
https://videocardz.com/newz/msi-changes-course-mag-qd-oled-monitors-will-get-firmware-updates-after-all

 

 

Here's more from MSI.... How long is it since Nvidia released 4060 and 4070? MSI gamers have been living in a lie. No one at MSI tech team/support have said that there is something wrong with the platform and RTX 4000 series cards. 

 

MSI fixes instability woes with NVIDIA RTX 4000 GPUs in new AMD motherboard BIOS update

Read more: https://www.tweaktown.com/news/99042/msi-fixes-instability-woes-with-nvidia-rtx-4000-gpus-in-new-amd-motherboard-bios-update/index.html

 

MSI says its new BIOS incorporates a few improvements in terms of compatibility with RTX 4000 products, without giving any details, although it does note that whatever work is applied, it especially pertains to the RTX 4060 and 4070 - bringing in cures for some instability issues.

 

So, if your RTX 4060 or 4070 graphics card has been playing up in an MSI 600 series motherboard, this update could be an absolute must-have to get things running smoothly again.