Presenting my new invention - RAMGUARD

[freeagent]

There was a guy on the toob a few years ago who 3D printed a bunch of ducts for his case. I could see that working ok if you are dead set on silence. Me? I don't care. My fans are running at full speed right now.

 

[Shrek]

I go the other way round, the more the fans the less the noise (as they can run slow)

 

[BoggledBeagle]

Time for ChipsetGUARD and SSDGUARD. I can invest in the project.

Or perhaps the blower style GPUs could return instead? I simply do not believe that a 3 slot blower card could not work well.

 

[freeagent]

I am glad blowers are gone, good riddance. Everyone seems to forget just how loud they really were. And how ineffective they were compared to what we have now. The real problem is that some GPU's are approaching 1000w. Have fun with thermals.

 

[lexluthermiester]

I am glad blowers are gone, good riddance.

Not me. Blowers have a place. Properly engineered they are very effective and you rarely know they're there. I'd like to see some of those triple slot video cards using double thick blowers pushing air through a wider fin stack and out the back of the case.

 

[freeagent]

Not me. Blowers have a place. Properly engineered they are very effective and you rarely know they're there. I'd like to see some of those triple slot video cards using double thick blowers pushing air through a wider fin stack and out the back of the case.

My 8800GTX was the last blower card I had.. that mofo got a Zalman

I may be out of touch with current offerings..

 

[lexluthermiester]

I may be out of touch with current offerings..

The key is proper engineering. Bad designs produce louder blowers that are less effective, which is what gives them a bad reputation. Good blowers kick ass.

 

[Waldorf]

@BoggledBeagle
looks nice and cool if you wanted to spend time on this, but i just switched to a LC loop, so i will never have to worry about any case/mb/ram temps (under full load) ever

 

[BoggledBeagle]

I am glad blowers are gone, good riddance. Everyone seems to forget just how loud they really were. And how ineffective they were compared to what we have now.

I think all of those cards were still small 2 slot cards. Today we accept that the cards can be larger and perhaps we know how to make better vapour chambers. Add some refinement of the blower itself and I think that the cards could just work fine and not dump all the hot air in the case.

 

[Teddy Bear]

The key is proper engineering. Bad designs produce louder blowers that are less effective, which is what gives them a bad reputation. Good blowers kick ass.

Years ago I had four EVGA GTX 970 Superclocked blowers in Quad SLI - That's where good blowers kick ass!

RIP EVGA
RIP SLI

I think all of those cards were still small 2 slot cards. Today we accept that the cards can be larger and perhaps we know how to make better vapour chambers. Add some refinement of the blower itself and I think that the cards could just work fine and not dump all the hot air in the case.

Part of today's heat control also has to do with innovation in case construction, fan construction, and airflow consideration in said design.

Blowers are still very much alive in the professional sector - If it was still big in the mainstream consumer market, there definitely will be major improvements and refinements over time.

I'd wager modern, refined blowers would look more tasteful and sleek than what we're used to now...

 

[A Computer Guy]

The key is proper engineering. Bad designs produce louder blowers that are less effective, which is what gives them a bad reputation. Good blowers kick ass.

My reference RX 5700 blower worked pretty well and the noise wasn't too bad but after 6 or so months it got loud because of garbage thermal pads on the GPU that started disintegrating causing the cooling to be less effective and the blower to ramp to max speed often.

 

[lexluthermiester]

My reference RX 5700 blower worked pretty well and the noise wasn't too bad but after 6 or so months it got loud because of garbage thermal pads on the GPU that started disintegrating causing the cooling to be less effective and the blower to ramp to max speed often.

If you haven't replaced those pads, doing so would likely return that card to proper operations.

 

[Space Lynx]

I'm glad someone else noticed the ram overheats from the GPU exhaust. Been saying this since the RTx. 30 series (even wrote it my reviews). Depending on the case I either put cardboard in-between or a fan on top of the GPU pointing at the ram.

I asked Nvidia about it and they said I'm the only one that has this problem lol. That was with 350w GPUs. Now we are at 600

what is considered a safe ram temperature for your average ddr5 6000 cl30 ram? i know ddr4 b-die it was considered to be around 50-55 celsius if I remember right? or was it 45 celsius before stability issues happened? I think DDR5 can go higher in temps without stability issue though?

fyi, my ram never gets past like 38 celsius when gaming, 6000 cl 30

 

[A Computer Guy]

If you haven't replaced those pads, doing so would likely return that card to proper operations.

I solved it by putting a waterblock on it. (I forgot to mention I got that card a bit more than 5 years ago.)

 

[Teddy Bear]

what is considered a safe ram temperature for your average ddr5 6000 cl30 ram? i know ddr4 b-die it was considered to be around 50-55 celsius if I remember right? or was it 45 celsius before stability issues happened? I think DDR5 can go higher in temps without stability issue though?

fyi, my ram never gets past like 38 celsius when gaming, 6000 cl 30

General idea to cover the million different types of heatspreader design and voltage/freq configurations:

(Without an auxiliary fan dedicated to cooling the DIMMs)

Limit ~105C JEDEC Specifications for tolerance (debatable)

Above 50c - Some dies may be stretching their stability
Below 50c - Fine
Below 45c - Better
Below 40c - You're either at idle or you have great airflow/heatspreaders!

Most non-low profile kits have decent heatspreaders so it should't be a concern running at rated speeds (EXPO/XMP)

6000 cl30 kits can run relatively low voltage for that speed - the faster kits need more voltage and the constant electrical load amplifies the temperature very quickly

DDR5 is more picky with temps than DDR4, but it is mostly resilient enough to run well at it's rated speeds near 50c - Overclocking is a can of worms that doesn't enjoy high tempratures

 

[kilo]

I'm not sure which country you're in, but if it's the USA, you're in luck as the three historical RAMGUARD trademarks are expired/abandoned.

In the UK, looks like it's never been registered. However DRAMGUARD is. And most RAM is DRAM...

New Zealand and Australia do have active RAMGUARD trademarks.

 

[ir_cow]

what is considered a safe ram temperature for your average ddr5 6000 cl30 ram? i know ddr4 b-die it was considered to be around 50-55 celsius if I remember right? or was it 45 celsius before stability issues happened? I think DDR5 can go higher in temps without stability issue though?

fyi, my ram never gets past like 38 celsius when gaming, 6000 cl 30

JEDEC is 85c for consumers. Some enterprise dram can be 100c+.

So the ICs can handle high temps without damage, but you will get errors at higher speeds. It's all safe, just when it will error out depends on mainly CAS, temperature, dram frequency and bank refresh (tRFC/tREFi).

 

[Teddy Bear]

JEDEC is 85c for consumers. Some enterprise dram can be 100c+.

So the ICs can handle high temps without damage, but you will get errors at higher speeds. It's all safe, just when it will error out depends on mainly CAS, temperature, dram frequency and bank refresh (tRFC/tREFi).

^^ This. Thanks, I glossed over the consumer spec, my bad.

tREFi is particularly sensitive to DIMMtemp on Hynix A-Die (usually found in kits with tight timings and caps out at 8200MT with good bins)

 

[BoggledBeagle]

I'm not sure which country you're in, but if it's the USA, you're in luck as the three historical RAMGUARD trademarks are expired/abandoned.
....

I do not intend to register the name or start selling it. I just quickly made up a name that explains what it does.

BTW GPUs could come with a sheet of paper or plastic somehow precreased to be easilly folded in shape to protect the motherboard and RAM sticks from the hot air instead of some stupid accessories as stickers, etc. But not everybody would use it and it would create a lot of unwanted waste. On the other hand, that is what sheets of stickers do as well.

 

[lexluthermiester]

I solved it by putting a waterblock on it. (I forgot to mention I got that card a bit more than 5 years ago.)

Ah, that works. Nice!

JEDEC is 85c for consumers. Some enterprise dram can be 100c+.

So the ICs can handle high temps without damage, but you will get errors at higher speeds. It's all safe, just when it will error out depends on mainly CAS, temperature, dram frequency and bank refresh (tRFC/tREFi).

Solid info. However, just because it can run that hot doesn't not mean we should let it ever get that hot. Sub 65C is optimal.

 

[freeagent]

I turned down my ram and played some games at the settings he asked for..

I would say it is totally safe lol..

 

[ir_cow]

I usually get errors in the 60-65c range with XMP. Though this isn't universal. CAS 30 may not being able to take 65c, but CAS 40 can. Unless someone wants to chart all speeds, all CAS for all ICs, it is just a recommendation to say below 60.

 

[lexluthermiester]

it is just a recommendation to say below 60.

I'll go along with that! The cooler the better!