Mr. Fox

The reason shunting works is because it deceives the GPU/firmware into believing it is pulling fewer watts than it actually is. Under full tilt load the GPU is pulling up to double what you see in GPU-Z or HWiNFO and I think you will find the WireView will confirm that. Voltage will also stay higher with the GPU pulling up twice the watts that it thinks it is using, although the voltage cap will not increase. It won't try to moderate itself because it has no idea what you're doing to it and won't attempt to decrease the voltage. Your software monitoring tools will become totally worthless with respect to GPU watts and show bogus information, but the core voltage reading should be as accurate as it was (or wasn't) prior to shunting.

Yeah, he is focusing on what matters most (family) and we probably won't see a KP card until next year at the soonest. He said as much. But, I think he could still drop them a few tidbits of useful information that would be actionable while they wait for him. For example, he could say something like "if you want to build a GPU that spanks everything else, just do A and B and you'll have something better than the cookie-cutter trash produced by your loser competitors."

Probably Vince's influence. If a PNY K|NGP|N 5090 (or 6090) drops I will be in. I'd feel like I was actually getting something worth having rather than paying a stupid price for an A$$zeus A$$troll 5090.

I buy everything from Thermal-Grizzly directly. They usually deliver from Germany in 3 to 5 days using FedEx or DHL. Did they send you the wrong one? ASUS uses the upside-down version. I believe they are the only ones and everyone else uses "normal" connections with the clip opposite the sense wire pins. ASUS have to be special (weirdos) and put the clip on the same side as the sense wire pins. I think their GPUs also burn more often than other brands, LOL. Nice you don't have to return it. Now you have a spare for either an ASUS or a normal GPU 12VHPWR connection. The opposite of what you need now, LOL.

Thank you. That makes perfect sense now where you placed it so the connector can shed its heat onto the backplate. Yikes. That is just too stupid for words, LOL. As they say in the deep south, "well, bless their heart" (which is not a kind thing to say). The cliché expression "run Forrest, run" is fitting here. I have CCD0 set for 55x and CCD1 set for 54x. I can't seem to get the ASUS hybrid overclock thing to work right. It's works pretty well on the AORUS Master, but not the Gene. PBO will go up to like 58x or 59x on a few cores and that 400 MHz difference is enough to be measurable in a game that only uses a few cores and threads at a time. Since it is not benching competition it doesn't matter much unless the FPS difference is too low for a good gaming experience.

Awesome. I really like them. In fact, just about anything Roman makes or particpates in meets my approval. Really smart guy, and apparently pretty financially well-off. The WireView doesn't fit every application and not everyone can use it. Too much variance in connector placement and heatsink design. It just barely works with the Bykski 4090 block and I will have to find out if it will fit with the Core waterblock. In fact, I had to file down a sharp corner on the Bykski block acrylic adjacent to the 12VHWPR so it could properly plug into the socket and latch. That would have been a melt due to improper seating had I not done that. I also prefer the round 12VHPWR cable like the one that came with my TT GF3 1650W PSU. The flat cables look better but they are not friendly whenever lateral curves are needed. With a flat cable bending up and down is fine. Left and right, not so much. Hard on the cable and connector and frustrating for the PC owner.

I use a WireView on my 4090 Suprim and the 4090 GB GPU that I sold. I have never felt the cable or the wireview feel warm to the touch using the WireView. I think it acts as a heat sink as well. Absolutely. The Green Goblin is so obstinate and unwilling to apologize or admit when they make mistakes that they continue pushing the use of this piece of crap connector type. Another example, among many, of a product manufacturer not caring about their products or the people that purchase them.

Can you explain or photo what you mean? I am trying to picture this and can't imagine where you applied the putty.

Yes it's at 100% in just about any game. In addition to that X3D thing I do not use PBO.

I sure do like having the tried and true 8-pin x 3 PCIe power cables instead of the chintzy 12VHPWR (aka 2x6 fire-starter) cables. I definitely got the memory temps down to an acceptable level with that last change. Hotspot temps are status quo for Radeon video cards. This game sucks, but the benchmark sure does use a lot more system memory and CPU resources than most. Right on the heels of 4090 performance in this title. Very acceptable price to performance at half the cost and even decent at 4K. GPU boost clock constantly between 3200 and 3300 MHz with no core offset.

Hopefully there will be a waterblock for this GPU. I've tried PTM pad, KPx and MX-6 and changed from putty to thermal pads and the temperatures are the same regardless. Memory junction temps are either too high or the sensors are off. 90°C memory temps reminds me of the 3090 when cryptomining and the 6900 XT (which hit 90°C on air or on water). The GPU is boosting to like 3400 on core with -100mV core voltage offset, no core clock offset and default (100%) power limit. Still impressive 4K results for a $700 GPU. Edit: so, I got the memory temps down 10°C but hotspot went up a few degrees. I used a thicker pad on the memory chips with higher w/mK rating. The others were making contact, but not being pressed enough to smash the thermal pad hard between the heatsink and chip. I used 0.5mm pads on everything. VRMs and inductors were getting pressed very hard but not the memory, so I replaced the memory thermal pads with 1.0mm 13w/mK pads. So, I am guessing the contact pressure on the die is not as great with the thicker pads on the memory and the pressure on the memory chips is much higher. I may try replacing the screws with springs with regular screws without springs to increase pressure on the die. Definitely an improvement even though hotspot temps increased slightly.

Flagships are made more attractive by deliberately suppressing and diminishing the performance and value of sub-flagship alternative products. The Green Goblin has already been deliberate about this by restricting PCIe lanes, PCIe bandwidth, memory bandwidth and memory capacity and deliberately castrating those products to make them less desirable. The performance differences are partly real and partly artificial and fabricated. They make the best product that everyone wants absurdly expensive and everything else undesirable and still overpriced. Don't want to pay out the kazoo for our best and most overpriced product? No problem you can spend less on this piece of crap that sells for twice what it is worth instead of four or five times more than it is worth. Just as allowing no good deed to go unpunished is a great way to discourage the continuance of good behavior, rewarding bad behavior effectively guarantees that more will follow. That would suck. Finding a silver lining, we would save money and more closely resemble the "good old days" when you could still beat the latest and greatest with something one or two generations older just by overclocking the crap out of it. The way things are trending, they'll probably block that as much as possible with the evil dictatorship approach to tech.

Yeah, the diminishing return when buying flagship products has always been a thing and should be expected. The absolutely ludicrous price gap, and the increase in all models across the board, has never been so over the top and can only be viewed as abusive and nefarious in both nature and scope in the way things are now. The approach they are taking doesn't deserve the consumer cooperation and validation is it receiving, so we are collectively cutting our own throat instead of cutting theirs. They deserve to be financially devastated for their behavior and we just keep giving them reasons to keep doing it and giving them our stamp of approval for them to get worse.

That's where I get stuck. The gap in performance is much smaller than the gap in price. If it is not 4.5 times more powerful it should not be 4.5 times more expensive. If it is I feel like the joke is on me and I'm their little money-making whore.

Pretty much, yup. Although mine is a fancy older model (2005 Ford Five Hundred SEL AWD) that I bought with like 50K miles on it. And, I drive everything until there is nothing left worth driving and the parts that are worn out cost more than the car is worth. You can get really nice cars for pennies on the dollar after all of the suckers and chumps have taken their losses. When it is worn out (another 200K miles to go) and not worth fixing I will sell it for what I paid for it. Lather, rinse, repeat.

We would probably see the reason why nobody has that except for the chosen ones. Just the thought of some of them owning a stock 5090 it's kind of scary. But not as scary as the thought that a desirable non-FE 5090 model costs more than what I paid for the car I've been driving for the last 5 years and will continue driving for at least a couple more.

I have sold several dead Intel CPUs on eBay for over $100 each, LOL. And, with very clear disclosures.

I agree, but I see it as a "corporate world view" that is not limited to America. This is a worldwide problem with big business and China is the biggest offender. Although, American corporations should be an exception to the norm. We should expect them to rise above and be better than that.

This design made it very hard for me to embrace Ryzen and I still really hate the design. It is an idiotic approach to processor design. It adds latency, unnecessary complexity and additional opportunities for failure. Along with the foolish elimination of hyperthreading the tiled design (a conceptually similar display of poor judgment and bad engineering decisions) was also a major factor in my rejection of Core Ultra and movement to Ryzen. The good old days are gone boys. What we're left with is the machinations of fundamentally bad people, educated idiots, greed on steroids and mainstream stupidity. End users get whatever end users get and they expect us all to feel grateful for getting screwed. Just as Big Pharma doesn't care about people getting better as long as they keep taking their lifelong poison meds, Big Tech doesn't care what is best for the people that buy their crap as long as the money keeps coming. They only pretend to care and listen when it is necessary to recover from a slowdown in revenue generation.

I have a Byski block on it already. That's the one in the photo. I ordered a Core for it (should be here Thursday) just because I think it looks a lot nicer. It's really chunky looking and has the backplate formed around the perimeter so there is no visible PCB or air space between the acrylic front and metal backplate. But, thank you for asking. What block for the Suprim do you have?

Sorry to hear about your CPU Brother @jaybee83. Something similar happened to Brother @Raiderman. I delidded his 9950X and tested it out. Everything worked great. I use the TG Mycro direct die block. He had issues with contact and thermals using a die frame and regular waterblock. He removed the CPU and when he put it back in the thing was dead. It has nothing visually wrong with it. By the way. I finally found a 48GB EXPO 8000 kit and it is totally stable in the Gigabyte Master. It's on the QVL as well. I believe it is a new SKU. All other 48GB memory I have tested on every AM5 motherboard was unstable. G.SKILL and TeamGroup Intel XMP kits were never stable even with all manual tuning. https://www.amazon.com/gp/product/B0DGRLW7V7 Nice to see you Brother @Rage Set. I hope all has been well with you. About the storage addiction... no explanation for it. Don't need it. Don't use it all. But I like doing things in excess. Congrats on the Astral Brother @johnksss. I'd be happy to delid the 14900KS for you. I think I may still have an extra LGA1700 IceMan or TG Mycro direct die block. If I still have one its yours as a gift. I am almost done with the projects. My tiny office has never felt so big, LOL. I still need to install the IceMan dual D5 pump in the O11D XL EVO. I just ran out of time so I slapped a single Byksi pump along with the Alphacool pump/res. I am going to have to fabricate a bracket for the dual D5 pump and just didn't have enough time or energy left to fuss with it right now. I also still need to delid the second 9950X CPU and install the second TG Mycro block. I'll probably do all of that at once in a couple of weeks after the Alpahcool Core block for the 4090 Suprim gets here. So, the O11D XL EVO turns out to be too small for push/pull with the fat radiators I installed. Heatkiller rads in top and bottom, Alphacool X-Flow in the side next to the mobo. Using the awesome Montech Metal Pro fans. I really like these fans. I have some in the Gene build as well. Since I have extras due to not being able to do push/pull on the O11D XL EVO I am going to use them to replace the inferior fans in the Gene. https://www.amazon.com/dp/B0DJ45JWL1?th=1

The cancer has caused a lot of reflection to occur. I feel mostly normal and haven't changed much overall, but after having to "donate" a $10K after-tax bonus from work to pay medical bills and further enrich a wealthy institution instead of stuffing most of it into a Roth IRA made me begin to re-evaluate things even more than the condition itself. The cynical side of my personality has been boosted rather dramatically, especially when it comes to things where the price tag is ludicrous, opportunistic and abusive and/or the product is a poor price to value proposition. I can't identify a better example of that than a 5090. I go in on 5/30 for tests to confirm the treatment was effective and boost my medical expense account balance from zero to whatever my deductible and copay is going to amount to. Side benefit... my updated forum signature is smaller now.

I really like the 9070 XT. First AMD GPU I have owned that I can say that. Not the world's fastest, but probably the best bang for the buck. I will try the newer driver this weekend and see how it compares. I just got the build with the Gene redone. I like it. The Antec Constellation C8 case is such a nice one, just a bit smaller than I would like it to be. After putting the 9070 XT in it and putting the RAM back on air, the dual 360 loop with only the CPU on water really cleaned things up. Now I need to redo the build with the triple radiator in the O11D XL EVO, delid the CPU in the Gene, then I think I am done. On a personal note, I am so happy I cancelled my Apex order and did not buy a 5090. I really like the AORUS Master mobo. A couple of minor gripes: no BIOS option to disable WiFi/BT M.2 slot and no setting to adjust the POST wait time to enter setup, but the Cabron was also missing the latter for some stupid reason. I will either physically remove the WiFi7/BT card that I have no use for or maybe see if I can mod the BIOS to add the two missing menu options. (Already asked Gigabyte and they just said no and used the excuse that nobody else is complaining about it and suggested I disable it in Device Manager if I don't want it. Losers.) After selling the Gigabyte 4090, white Apex, TG Xtreem 8200 and 14900KS I purchased the second 9950X, second TG Mycro, 5 radiators, 12 fans, my third Sabrent Quad NVMe card, two D5 pump/reservoirs, tubing, fittings, an Alphacool Core block for the 4090 Suprim (replacing the Bykski block primarily for aesthetic reasons favoring the look of the Core) a couple more 2TB NVMe, a 9600X CPU for the B850 build given to my grand daughters, two 800W UPSes and still have $600 sitting in my PayPal account, LOL. I would have needed to add money to buy a non-FE 5090. When I stop and think about this it makes the thought of buying one just seems totally ludicrous to me now. I was very tempted to blow $3K+ on a GPU I didn't need.