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

[ryan]

if you blow up the images you get a better idea. but yes I guess its not a accurate comparison but it does look good neverless.

 

8gb? I wish my 3060 was 8gb its a paltry 6gb and usually only low textures fly.

 

also got my BE200 wifi adaptor. 6ghz band showed up with default settings nothing changed, just installed drivers and 6ghz band is showing. I guess ill be buying 3 more of them. does 5800mbps theoretically.

[electrosoft]

50 minutes ago, tps3443 said:

I considered this about the 4060, however it is only 8GB which further hurts its performance and 1% lows especially. Making it a useless GPU for newly launched titles as most aren’t optimized all that great. If I’m not mistake though I think game devs are hesitant to optimize for 8GB VRAM. Apparently it’s too time consuming. Check these 1% lows on 8GB 4060. So in this case even with a 4060 having Frame Gen, It still won’t save it. 
 

 

 

 

 

 

In the end, frames rule the game. Remember, I had a 3060 12GB before and all the VRAM in the world will not save you from a GPU that just isn't up to the task. Pumping up details to take advantage of the 12GB = steadily decreasing frame rates and it tanked at 1080p high on Hogwarts, WoW and Fallout. 3070 8GB ran circles around it. Yes you have a 50% larger VRAM buffer but now you need more GPU power to handle the higher level of detail/settings. Chicken or the egg? 🙂

 

 

On the other hand, you can dial down the details to the point you can even limp by at 4k on a 3060 which I did actually when I was waiting for my "next card" after selling my 3090 FE and before I got my KPE 3090. That one was actually an 8GB 3060 and I ran it at 4k low details for over a month. Such low settings = no VRAM issues and less stress on the GPU.

 

Ideally, you would want a 4060ti 16GB and you get the best of both worlds. An even larger VRAM buffer AND FG that will absolutely shine at 1080p but Nvidia is stupid with their pricing and you're looking at a $150 difference but it makes the most sense for 1080p gaming. Albeit I do see the value proposition with the 3060 12GB. It's a good starter card and overall pretty solid at 1080p.

 

Then again my nephew is loving his A380 6GB GPU so perspective I guess. 🙂

 

 

 

 

 

[ryan]

yeah perspective plays a large role, Alot of people swear by ultra settings but I dont notice the difference. I run low/medium textures and everything else cranked because why not in grid legends at 4k and get a locked 60fps with underclocking and undervolting cpu/gpu. alot of newer games are not even that demanding. hogwarts and hell divers sure ill get 30fps all low but if I remember correctly they have dlss which if you sit back a few feet is a game changer.

 

overall I say the 3060 is perfect for 1080p/1440p gaming and should last a deal longer, I just hope that nvidia releases the new ssd as ram setting for older cards, they are talking about using m.2 ssds storage as vram

 

 

maybe your son can beat this score but for a 115w 3060 in the omen 16 it's bonkers. quite a bit higher than all other omen 16 benchmarks.

 

my highest score yet, I really did not think using a preview version of windows 11 and 20 dollar paste would make a 300 point uplift in timespy. I think my previous high was around 9550-9600.

 

 

[chew]

nothing fancy just rank 6 worldwide memory frequency but i no longer play the hwbot game.,,,,,,purely for fun.

 

https://valid.x86.fr/fzl0a8

[Papusan]

 

On 3/31/2024 at 11:42 AM, Mr. Fox said:

Happy Easter, everyone. He is risen!

 

Happy Easter holidays my friends. I'm..... Halfway back🙂 

 

 

And I hope Evga can as well... But this won't save the consumers screwed's  data. Backup bacup backup. Can't say it loud enough.

EVGA promises to replace burned hard drives

The power supply manufacturer crawls to the cross.

Recently we wrote about Reddit user Sgircys who received a new power supply under warranty from the manufacturer EVGA , who had failed to inform that the pin layout of the power connectors had changed. The result was that Sgircys, using their existing cables, sent too much voltage to their SATA drives, destroying them.

 

On 3/23/2024 at 8:52 AM, Rage Set said:

How the mighty have fallen. This news story is bad for the legacy of EVGA if true. I guess if all the good employees left, the shell company doesn't care.

 

 

 

On 3/30/2024 at 12:02 PM, electrosoft said:

The refreshed models now have mini led options but I'm still not seeing anything on the horizon calling me.

 

Welcome to Papusan's world. Laptops ended for me in 2016. I never bought anything new after that year. Everything that come after stinked. The 870DM had a good life for me and is still. 6700K-7700K-8700K and finally 9900K in same laptop. Nothing can come close to that in the laptop world🙂 So good bye and.....

 

Btw. Me enjoy the anologe world. A hell lot better than the disgusting new modern bling bling all around, bloat and RGB world.

 

 

On 3/27/2024 at 4:05 PM, Rage Set said:

According to Steve in GN's latest news video, some members of the team that developed EVGA mobo BIOS left for Asrock....keep that in mind for future mobo purchases. 

Hmmm. Splave is still on the old Z690 and thats that. Say no more.. On a binned Z690 MB. Asrock failed with Z790. Of course they need better MB engineers. That's life. What a fail of MSI not grab up those guys. For the records.... MSI have still not pushed hard their main OC sku.... Lightning. 

 

What a failure from a big brand as MSI. They didn't see they risked be only a gamer boy brand? 

[Mr. Fox]

2 hours ago, Papusan said:

Btw. Me enjoy the anologe world. A hell lot better than the disgusting new modern bling bling all around, bloat and RGB world.

 

Welcome back.

 

Cool analog gauges. Would be nice to have them mounted in a combined console instead of separate. Would be cool attached to the bottom or top of a monitor. It looks like the lighting changes from green to red based on the temperatue or load. Other than that insanely overpriced EVGA open frame tubular bench I don't think I have seen gauges for PC like these before. I like them, too. Very nice.

[Papusan]

1 hour ago, Mr. Fox said:

Welcome back.

 

Cool analog gauges. Would be nice to have them mounted in a combined console instead of separate. Would be cool attached to the bottom or top of a monitor. It looks like the lighting changes from green to red based on the temperatue or load. Other than that insanely overpriced EVGA open frame tubular bench I don't think I have seen gauges for PC like these before. I like them, too. Very nice.

 

Hi thanks. I like that people/engineers think out of the box. Hate the ONLY ALL SHALL HAVE ---- RGB or Bling in every shape. Same also with Todays JOKEBOOKS. All must be the same. Disgusting!!!

 

Ps. Sent you an PM.

 

Btw. Cpu degradation.... Worth to see. Buth expect stupiditi from this stupid pc editor. Thats normal.

 

 

On 3/28/2024 at 4:16 AM, chew said:

chasing dreams.......19K multi mem score or cpu death......whichever comes first........

 

Fantastic score. But nope.... This is not the CPU death. And if it was.... AMD don't scale very well with sky high mem clocks. Still damn good score 🙂

[Mr. Fox]

4 hours ago, Papusan said:

Btw. Cpu degradation.... Worth to see. Buth expect stupiditi from this stupid pc editor. Thats normal.

 

While I don't agree with all that is being said, I do agree that CPU degration is generally due to absurd abusive behavior and not "normal" overclocked benching. Running long and repeated sessions of severely stressful things like Y-Cruncher, Furmark, OCCT and Prime95 to "prove stability" is just stupid and unecessary. Doing that can degrade CPUs and GPUs even if they are not overclocked. It makes no sense. Same can be said of running excessive amounts of SSD benchmarks. They are only covered under warranty for so many write operations and they wear out. All you are doing to chewing up the drive and memory. If you run an SSD benchmark, you know what your drive is capable of. No need (or benefit) to run it again and again. If your system passes a memory test without error and you haven't changed anything, no need to run it again and again unless a new problem surfaces that needs to be investigated. If your CPU can pass Cinebench and you can run all 3DMark benchmarks, and play games for hours without a hiccup, it is stable. No need for Prime95, OCCT and Furmark. All those do is prove that if you run abusive test routines your system won't be stable. And, it is not healthy for the parts to do that to them over and over again.

 

It is kind of like walking into a room and you start screaming at a person for no reason. You push him around and call him names, maybe punch him and keep screaming at him. If he doesn't react or he remains pleasant then he is stable. If he reacts in a negative way, screams back or returns the punch then he is unstable. If your assault hurts him physically or emotionally then he is defective. I don't think so. Any unpleasant reaction should come as no suprise. That is unwarranted abuse, and it is very different than challenging him to a race, a test of knowledge, or a contest of strength or skill. 

[electrosoft]

4 hours ago, Mr. Fox said:

While I don't agree with all that is being said, I do agree that CPU degration is generally due to absurd abusive behavior and not "normal" overclocked benching. Running long and repeated sessions of severely stressful things like Y-Cruncher, Furmark, OCCT and Prime95 to "prove stability" is just stupid and unecessary. Doing that can degrade CPUs and GPUs even if they are not overclocked. It makes no sense. Same can be said of running excessive amounts of SSD benchmarks. They are only covered under warranty for so many write operations and they wear out. All you are doing to chewing up the drive and memory. If you run an SSD benchmark, you know what your drive is capable of. No need (or benefit) to run it again and again. If your system passes a memory test without error and you haven't changed anything, no need to run it again and again unless a new problem surfaces that needs to be investigated. If your CPU can pass Cinebench and you can run all 3DMark benchmarks, and play games for hours without a hiccup, it is stable. No need for Prime95, OCCT and Furmark. All those do is prove that if you run abusive test routines your system won't be stable. And, it is not healthy for the parts to do that to them over and over again.

 

This and more of this and then a sprinkle of this on top.....

 

I've never understood that "grind it to the edge for hours or days" mentality.  I usually have a benching profile and a D2D profile. In no way, shape or form am I going to run my benching profile(s) under those types of conditions for hours on end. That is just asinine.

 

Worse thing you can do is listen to others try to tell you how to qualify your chip based on their criteria instead of your own. Don't think my chip meets your OC standards? Ah well, go cry in a corner. 🙂

 

 

 

 

[Clamibot]

I just encountered something very strange when overclocking my Lenovo Legion Go. I decided to try and overclock the iGPU to gain some frames in games where I'm just a little under my 144 fps target, and I discovered that overclocking the iGPU... decreases performance?

 

This makes no sense and I've never encountered this before. I've overclocked both Nvidia and AMD dGPUs and Intel iGPUs and never have I ever seen a performance DECREASE when overclocking a GPU, only increases. Does anyone know why something like this would happen? I'm not temperature or power limited as far as I can tell. There's lots of thermal headroom in this handheld to push it significantly past stock performance. I know there's a voltage limit of 1.2v for this particular iGPU, but I'm not hitting that limit either.

 

Any thoughts? I'm at a loss here. My Steam Deck doesn't do this and I can overclock its iGPU fine for those performance gains.

[ryan]

get ryzen controller, find out your cpus max tdp set and forget...should give you a 50 percent boost

 

this is with experience on my amd ryzen 7 5825u its tdp by default is 15w when i set all the tdp values to 25w(safe) it bumps performance by 43 percent

 

i was getting 900 stock in timespy..now im getting 1500

 

Ryzen™ Z1 Extreme Processor | AMD

 

you can super safely set yours to 30w and btw i set mine to 38w 13w over what the max is and its safe, just don't play with voltages unless your seanwee or talon

 

do a run stock the best you can get

 

download this

 

Ryzen Controller

 

set temp to 90c and all boxes that say tdp set three or four of them to 30w

 

run a test and report back

[Mr. Fox]

3 hours ago, Clamibot said:

I just encountered something very strange when overclocking my Lenovo Legion Go. I decided to try and overclock the iGPU to gain some frames in games where I'm just a little under my 144 fps target, and I discovered that overclocking the iGPU... decreases performance?

 

This makes no sense and I've never encountered this before. I've overclocked both Nvidia and AMD dGPUs and Intel iGPUs and never have I ever seen a performance DECREASE when overclocking a GPU, only increases. Does anyone know why something like this would happen? I'm not temperature or power limited as far as I can tell. There's lots of thermal headroom in this handheld to push it significantly past stock performance. I know there's a voltage limit of 1.2v for this particular iGPU, but I'm not hitting that limit either.

 

Any thoughts? I'm at a loss here. My Steam Deck doesn't do this and I can overclock its iGPU fine for those performance gains.

I suspect it has something to do with the platform rather than the CPU/iGPU. Lenovo probably put some kind of limiter in place. Maybe a power limit on the chip includes the CPU and iGPU combined? When you run benchmarks does the CPU behavior change? What about the Physics and Combined scores in something like Fire Strike? Do those go down with the iGPU overclocked? What does HWiNFO64 show for total CPU Package Power run both ways?

[ryan]

2nd place.

 

[Clamibot]

22 hours ago, ryan said:

get ryzen controller, find out your cpus max tdp set and forget...should give you a 50 percent boost

 

this is with experience on my amd ryzen 7 5825u its tdp by default is 15w when i set all the tdp values to 25w(safe) it bumps performance by 43 percent

 

i was getting 900 stock in timespy..now im getting 1500

 

Ryzen™ Z1 Extreme Processor | AMD

 

you can super safely set yours to 30w and btw i set mine to 38w 13w over what the max is and its safe, just don't play with voltages unless your seanwee or talon

 

do a run stock the best you can get

 

download this

 

Ryzen Controller

 

set temp to 90c and all boxes that say tdp set three or four of them to 30w

 

run a test and report back

 

22 hours ago, Mr. Fox said:

I suspect it has something to do with the platform rather than the CPU/iGPU. Lenovo probably put some kind of limiter in place. Maybe a power limit on the chip includes the CPU and iGPU combined? When you run benchmarks does the CPU behavior change? What about the Physics and Combined scores in something like Fire Strike? Do those go down with the iGPU overclocked? What does HWiNFO64 show for total CPU Package Power run both ways?

 

Benchmark scores go down with the overclock applied. I raised the power limit as high as I could go in the BIOS (54 watts). I even tried that Ryzen Controller application, but that just yielded the same results as using UXTU.

@Mr. Fox I think you're correct about there being some sort of limiter being in place. The system was not respecting my set power limits and refused to draw up to the 54 watt limit I set in the BIOS. I did some digging using UXTU and found the amp limits and temperature limits were set really low, so there was no way I was gonna reach my power target. Raising all of those to the maximum value I could set them to resulted in a performance increase.

I however still cannot push the GPU clocks past around 2.5 GHz without seeing a performance drop. Looks like some other sort of limiter is in place that I need to find and eliminate. Based on what I see, it looks like there's some really early thermal throttling going on despite temps being nowhere near the thermal throttle point.

 

Thanks for your help guys!

 

Edit: I think I found the culprit. There's a safety feature that will thermal throttle the APU if it reaches 80°C so the device won't burn your hands... except the device never gets hot enough to do that. I'll need to find a way to disable this to get the iGPU overclock working.

Supposedly all AMD CPUs are multiplier unlocked, but that doesn't help when the firmware sets stupid thermal limits.
 

[johnksss]

So I went ahead and tried my luck again and this time I got a SP105 and able to boot 9K memory. And no I did not start 9K memory testing, but 8800 is almost stable. Maybe bench stable, but not long term stable as of yet.

[Mr. Fox]

On 3/28/2024 at 7:55 PM, ryan said:

Anyone here use PTM 7950? I just ordered some and should be here tomorrow. going to try to break the 12600 record with the 11800h. Will be interesting to see how well they work, my gpu fan is running at 87rpm not sure if its built up dust or what but hopefully the portable duster works well enough and fixes some of the heating issues. 3060 is still going strong your son should be good for a while

 

On 3/29/2024 at 8:35 PM, Mr. Fox said:

Maybe one weekend when I don't have something that needs to be done I will tear apart my Precision 7720 and put some of the PTM 7950 termal paste I have on it. That's why I bought the PTM 7950 to begin with. I just haven't mustered enough desire to be bothered with the hassle. Whenever I need to use it I am always "impressed" by how hot it runs. I should put some on my wife's laptop (Eluktronics Mech 15). That sucker has always been a hot mess. It takes a lot for me to feel motivated to work on either one. The Mech 15 is much easier though. Just a few screws to remove the bottom cover and everything is accessible. But, it's demoralizing working on laptops to me now.

 

The Dell, of course, makes everything much more of a pain in the butt than it needs to be. Everything involves lots of unnecessary screws, cables and foolish rigmarole. Build quality 10/10. Intelligent, user-friendly design 1/10.

Well, that wasn't a great use of my time this evening. I must have already applied it before and forgot. I don't remember when, but I hardly ever use my laptop. Since I tore it apart I had to apply it again. Didn't make a huge difference... 3°C on CPU and 1°C on GPU. Still runs like a hot turdbook. But, I digress... it is a hot turdbook. Oh well. Que sera sera. Glad it's not that important. It does last longer than normal thermal paste does on a laptop, so I will give it that much. 

  

1 hour ago, johnksss said:

So I went ahead and tried my luck again and this time I got a SP105 and able to boot 9K memory. And no I did not start 9K memory testing, but 8800 is almost stable. Maybe bench stable, but not long term stable as of yet.

Glad I popped in here right before hitting the sack. Nice to see you, Brother John.

Looks like that one is a lot better than the other one. 

 

I got all my money back from NewEgg, Amazon (finally) and Central and decided to hang it up rather than keep rolling the dice. Four turds in a row was too much.

 

I've got good CPUs samples already and one incoming from Germany so I'm done rolling the dice in the lottery. From now on I will overpay a little bit for a used CPU that has SP ratings and a VF curve I can see before I buy. That seems to be the safer play.

 

I think I am going to skip a few generations now (famous last words). This seems like it is getting more expensive, like everything else, and it was already expensive before. Plus, I hardly ever bench anymore so I'm not having as much fun as frequently as I used to, which makes it even more expensive because I'm not getting the same amount of pleasure/use out of the process.

 

Have a good night.

3 hours ago, Clamibot said:

 

 

Benchmark scores go down with the overclock applied. I raised the power limit as high as I could go in the BIOS (54 watts). I even tried that Ryzen Controller application, but that just yielded the same results as using UXTU.

@Mr. Fox I think you're correct about there being some sort of limiter being in place. The system was not respecting my set power limits and refused to draw up to the 54 watt limit I set in the BIOS. I did some digging using UXTU and found the amp limits and temperature limits were set really low, so there was no way I was gonna reach my power target. Raising all of those to the maximum value I could set them to resulted in a performance increase.

I however still cannot push the GPU clocks past around 2.5 GHz without seeing a performance drop. Looks like some other sort of limiter is in place that I need to find and eliminate. Based on what I see, it looks like there's some really early thermal throttling going on despite temps being nowhere near the thermal throttle point.

 

Thanks for your help guys!

 

Edit: I think I found the culprit. There's a safety feature that will thermal throttle the APU if it reaches 80°C so the device won't burn your hands... except the device never gets hot enough to do that. I'll need to find a way to disable this to get the iGPU overclock working.

Glad I might have pointed you in the right direction. I hope you can figure it out.

3 hours ago, Clamibot said:

Supposedly all AMD CPUs are multiplier unlocked, but that doesn't help when the firmware sets stupid thermal limits.

Well... you wouldn't expect me to find something nice to say about AMD, but I don't mind casting stones at Intel either when they deserve it. That is a really huge problem on Intel-powered laptops as well. They've taken the castration to the next level on turdbooks. You don't own them, they own you. Our friend Joe (Trafficante, aka USMC362) from the old NBR forum bought a 14th Gen Intel turdbook and it is power-locked like a big dog. ThrottleStop performance cripple reasons light up like a Christmas Tree and there is nothing you can do to work around it.  It's unlocked and you can overclock the crap out of it while idling, but as soon as you put it under load the clocks fall, it hits the fused power limit and it runs like garbage. It just sucks. In part because it's an ASUS turdbook, but they all suck now.

[ryan]

Bro, My temps were great for a while but it could have been that my window was open and its 0c outside lol. I dont know windows are closed and instead of 87c max im seeing 95c max at lower clocks 4.4ghz vs 4.6ghz but score wise after de dusting and applying the thermal paste the performance on the gpu has gone up tremendously. I think its possible HP just did a terrible job with the stock paste and the temp differences im seeing are superficial. 

 

See this, the value speaks for itself. I have been posting benchmarks for 2 years and my old max of roughly 9650 and it was basically impossible to get higher as power was the limiting factor. now that I can undervolt and have lower temps its making a pretty big difference in overall score, if I break 10k It will be like getting down on one knee to this old omen. I will be incredibly happy about it and it's looking like its a possibility

 

 

 

I think its not impressive for a desktop 3060 or a full powered 3060 in laptops with decent cooling stock. like the score I was getting 8300 was the old stock and the old stock overclock score of roughly 9000 was hard to pass and a long shot. the tdp increase with the new vbios and the iets gt500 and ptm7950 all add up to a big gain. in games the difference is larger, especially at 4k.

[johnksss]

38 minutes ago, Mr. Fox said:

 

Glad I popped in here right before hitting the sack. Nice to see you, Brother John.

Looks like that one is a lot better than the other one. 

 

I got all my money back from NewEgg, Amazon (finally) and Central and decided to hang it up rather than keep rolling the dice. Four turds in a row was too much.

 

I've got good CPUs samples already and one incoming from Germany so I'm done rolling the dice in the lottery. From now on I will overpay a little bit for a used CPU that has SP ratings and a VF curve I can see before I buy. That seems to be the safer play.

 

I think I am going to skip a few generations now (famous last words). This seems like it is getting more expensive, like everything else, and it was already expensive before. Plus, I hardly ever bench anymore so I'm not having as much fun as frequently as I used to, which makes it even more expensive because I'm not getting the same amount of pleasure/use out of the process.

 

Have a good night.

 

Hey there Brother Fox.

This one does seem a bit better, that's for sure. I may just ride this one out and get rid of my 13900KS.

I only rolled the dice this time because I had to go get some parts for a customer and was like....What do you have close to a x346 or higher. They came back with this one. And it only cost me like a 100 bucks to swap cpu's.

 

I know what you mean. I haven't been benching much at all. Been to busy with work related stuff.

[Mr. Fox]

I have a couple of these cheap $32.50 4TB NVMe out for delivery today. It will be interesting to see if they are good or garbage. They look like counterfeit Samsung NVMe but I don't care what brand, fake or not, as long as they work as intended. https://www.aliexpress.us/item/3256806361905543.html 

  

7 hours ago, ryan said:

in games the difference is larger, especially at 4k.

I have been rocking a high-end ASUS 4K 144Hz monitor for a good while now and while it produces gorgeous images and games look super nice on it, I am still having a very difficult time liking 4K for normal daily use. The scaling sucks with any setting other than 100% and the rendering of text leaves a lot to be desired. Applications like CPU-Z and GPU-Z with lots of text are especially ugly-looking. Productivity apps (Word, Excel, PowerPoint, etc.) and web browsing look like pure crap at any scaling other than 100% and some things are just too small to be a pleasant reading experience at 100% scaling. IMHO, 2560x1440 is still the all-around best resolution for everything. Just set it to 100% scaling and everything is perfect-looking all the time. I always hated 4K laptop screens (and still do) because they are pretty miserable products to use, but even a 27-inch 4K screen a couple of feet away from my face is less pleasant to look at than 1440p.

 

[Reciever]

54 minutes ago, Mr. Fox said:

I have a couple of these cheap $32.50 4TB NVMe out for delivery today. It will be interesting to see if they are good or garbage. They look like counterfeit Samsung NVMe but I don't care what brand, fake or not, as long as they work as intended. https://www.aliexpress.us/item/3256806361905543.html 

  

I have been rocking a high-end ASUS 4K 144Hz monitor for a good while now and while it produces gorgeous images and games look super nice on it, I am still having a very difficult time liking 4K for normal daily use. The scaling sucks with any setting other than 100% and the rendering of text leaves a lot to be desired. Applications like CPU-Z and GPU-Z with lots of text are especially ugly-looking. Productivity apps (Word, Excel, PowerPoint, etc.) and web browsing look like pure crap at any scaling other than 100% and some things are just too small to be a pleasant reading experience at 100% scaling. IMHO, 2560x1440 is still the all-around best resolution for everything. Just set it to 100% scaling and everything is perfect-looking all the time. I always hated 4K laptop screens (and still do) because they are pretty miserable products to use, but even a 27-inch 4K screen a couple of feet away from my face is less pleasant to look at than 1440p.

 

 

Very interested to see if those work out as I have a workstation/server that would love the SSD treatment for the VM's. Would make my life so much easier in terms of cables too lol. Even if its SATA based and not NVME, still would be worthwhile if the storage is actually there.

 

I agree also with your perspective on resolution, at 27" 2560x1440p @ 100% is very comfortable. On my 4K120hz TV I still use 125% and thats at 42" probably around 3-5Ft away pending how I am using the screen.

[tps3443]

2 hours ago, Mr. Fox said:

I have a couple of these cheap $32.50 4TB NVMe out for delivery today. It will be interesting to see if they are good or garbage. They look like counterfeit Samsung NVMe but I don't care what brand, fake or not, as long as they work as intended. https://www.aliexpress.us/item/3256806361905543.html 

  

I have been rocking a high-end ASUS 4K 144Hz monitor for a good while now and while it produces gorgeous images and games look super nice on it, I am still having a very difficult time liking 4K for normal daily use. The scaling sucks with any setting other than 100% and the rendering of text leaves a lot to be desired. Applications like CPU-Z and GPU-Z with lots of text are especially ugly-looking. Productivity apps (Word, Excel, PowerPoint, etc.) and web browsing look like pure crap at any scaling other than 100% and some things are just too small to be a pleasant reading experience at 100% scaling. IMHO, 2560x1440 is still the all-around best resolution for everything. Just set it to 100% scaling and everything is perfect-looking all the time. I always hated 4K laptop screens (and still do) because they are pretty miserable products to use, but even a 27-inch 4K screen a couple of feet away from my face is less pleasant to look at than 1440p.

 

I have found that 4K is definitely best at 100% scaling with windows. However, while 27” is very sharp and high PPI it is too small to reliably and quickly use the monitor while reading text etc. 

 

This was why I moved up to a 48” OLED 4K Monitor. I think probably 42” would be good as well too though.
 

48 inches is pretty large. But sitting 3ft-4ft away you can read any text like it’s nothing and actually use the panel very easily. I can slide back to 8ft and still read the text at 100% scaling. I think using smaller 4K panels like 27” will make our brains and eyes work much harder to read and manage apps. 

[Papusan]

4 hours ago, Mr. Fox said:

I have a couple of these cheap $32.50 4TB NVMe out for delivery today. It will be interesting to see if they are good or garbage. They look like counterfeit Samsung NVMe but I don't care what brand, fake or not, as long as they work as intended. https://www.aliexpress.us/item/3256806361905543.html 

 

Have read about those before. I wouldn't trust my important data in that drives if you don't have an ongoing backup schedule running in the background. Samsung’s Magician Software can confirms that it's a counterfeit drive or not. The software always validates whether a drive is fake or not.

 

Edit. 

Fake Samsung 980 Pro SSDs Are Spreading Around

Mar 18, 2023 — The merchant sold fake Samsung 980 Pro 2TB SSDs on Xianyu, Taobao's second-hand market, for 880 yuan or $127.77

 

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

 

I'm still not in so very good shape. Hence I post less. I have barely touched an pc for soon 2 weeks. But managed to order 2x 14900KS. But with my luck and Intel's trash binning for the last unlocked chips for 1700 platform, I doub't I will see good chips. Maybe better than my SP94 14900K but not much more.

 

At least my money from previous 14900KS is on the way back. That's something.

 

17780 nok is $1660 USD. Hail to the Norwegian TaX system (greed). Or equal disgusting as Intels binning for this gen KS chips.

 

 

 

 

 

 

[tps3443]

I'm done binning on LGA1700. I almost ended up with a 14900KS. But just decided to just be happy with my R batch. It runs great at 6.1P/4.9E I have been running this speed for about a week now daily. And I have just forgotten all about it. 

Using a manual SVID of 1.395V with LLC5 in the bios. 

Still have not stepped on it with heavy voltages. I have no doubt that 6.2 can be done with 1.300+ load V-core. But, staying in the 1.200's load range is a little better long term. 



 

[johnksss]

Just wish this one had better E cores, but since I'm not really benching right now i'm good with it.

[Papusan]

14 minutes ago, tps3443 said:

I'm done binning on LGA1700. I almost ended up with a 14900KS. But just decided to just be happy with my R batch. It runs great at 6.1P/4.9E I have been running this speed for about a week now daily. And I have just forgotten all about it. 

Using a manual SVID of 1.395V with LLC5 in the bios. 

Still have not stepped on it with heavy voltages. I have no doubt that 6.2 can be done with 1.300+ load V-core. But, staying in the 1.200's load range is a little better long term. 



 

 

All 14900KS should be able to run it this way. Not this trash here. And be you sure... Probably many KS chips out there that's worse than this.

 

8 minutes ago, johnksss said:

Just wish this one had better E cores, but since I'm not really benching right now i'm good with it.

 

In short... Sad ending on the 1700 platform. Both the 13th gen K and the KS was a good upgrade. 

 

Good to see you bro John. And same for you all 🙂