Netac Kickstarts Research and Development Process for 10 GHz DDR5 Memory

[AsRock]

I thought Trek Technology invented the thumb drive, maybe i need to look in to it again it's been some time, don't remember it being and Chinese company.

Has this Chinese company been around 15+ years ?.

Even if they could pull i t off i would not support such BS.

 

[Tartaros]

Netac, a Chinese company based in Shenzen claiming to be the inventor of USB flash drive

 

[TranceHead]

Software isn't people.

Ive always used 1024×1024 ‍

 

[Frick]

Ive always used 1024×1024 ‍

Even outside how Windows looks at hard drives?

 

[Kaotik]

I thought Trek Technology invented the thumb drive, maybe i need to look in to it again it's been some time, don't remember it being and Chinese company.

Trek released first one for sale

Has this Chinese company been around 15+ years ?.

Yes, it was founded in 1999 and filed for the patent the same year.

Even if they could pull i t off i would not support such BS.

Call it BS or not, they still have the first patent for USB Flash Drive (or rather, now expired patent)

 

[AsRock]

And Israeli have the 1st denied one ?, as i understand it. Either way looks all kinda sketchy to me.

Who actually owns it now ?,

 

[mtcn77]

And Israeli have the 1st denied one ?, as i understand it. Either way looks all kinda sketchy to me.

Who actually owns it now ?,

I have to think there is an underlying story. You know, the inventor of the microchips that fostered the Intel 8008 got sued for millions of dollars. I bet many got discouraged from privately investing in their fields because of this reason. It should be condemned as a form of strong arming and terrorizing intellectual property developers by patent trolling practices making them comply to harsh treatment for no other reason than their own resourcefulness...

 

[Deleted member 205776]

But this memory will run extra hot simply because of all the spyware/backdoors/loggers etc that they will add in, which will run in the background, completely obscured from public view, which of course will be impossible to remove

You are being spied on more by the same companies that you trust than you will ever be by these random magical RAM sticks (who will definitely run at 10 GHz without melting).

 

[Prima.Vera]

10GHz For DDR5 is simmilar with 5000MHz -DD4 or 2500MHz - DDR3.
Really not that impressive, relax.

 

[Kaotik]

10GHz For DDR5 is simmilar with 5000MHz -DD4 or 2500MHz - DDR3.
Really not that impressive, relax.

Any one of those would be beyond impressive.
Your post is prime example of what kind of confusion Marketing BS in news causes.
Nothing in DDR3 works at 2500 MHz. Nothing in DDR4 works at 5000 MHz. Nothing in DDR5 will work at 10 GHz.
Also even if you had said DDR5-10000 = DDR4-5000 = DDR3-2500 you'd still be wrong, they're not even close to being similar with each other in anything but latency.

 

[ixi]

10GHz For DDR5 is simmilar with 5000MHz -DD4 or 2500MHz - DDR3.
Really not that impressive, relax.

Mathematics, stoink!

 

[zlobby]

CL40? Yeah, no.

 

[voltage]

I will laugh my ass off when I see the CAS rating.

 

[TumbleGeorge]

DDR base internal frequency steel is between 100-200MHz. Citated numbers in comments are effective data rate not frequency.

 

[IceShroom]

TPU is kind of spreading false information by saying the DDR5 is running at 10GHz. A 10GHz memory would have been good for new CPU, but sadly currently transistor based system only run upto 5.1-5.3 GHz safely. The headline should be - "Netac Kickstarts Research and Development Process for 10 GT/s DDR5 Memory" or "Netac Kickstarts Research and Development Process for 10 Gbps DDR5 Memory".

DDR or Double Data Rate memory sends 2-bit per clock. So actual RAM speed is half of data rate. And JEDEC usually rates their DDR chip with Data transfer rate, not by clock speed. This is from JEDEC : https://www.jedec.org/news/pressrel...rd-advancing-next-generation-high-performance

 

[Wirko]

CL40? Yeah, no.

Latency, in absolute terms (nanoseconds), is more or less constant over generations of dynamic RAM. There has been some progress, anyway. The Intel 2104A DRAM from 1976 had a 150 ns access time, you can check it here. That was for random access, hence the name "random access memory". Today's chips do the same thing three times faster.

 

[zlobby]

Latency, in absolute terms (nanoseconds), is more or less constant over generations of dynamic RAM. There has been some progress, anyway. The Intel 2104A DRAM from 1976 had a 150 ns access time, you can check it here. That was for random access, hence the name "random access memory". Today's chips do the same thing three times faster.

It's a relatively long story but the bottom line is that these MT/s are only good for bandwidth. As for latency we haven't moved a bit.

 

[Wirko]

It's a relatively long story but the bottom line is that these MT/s are only good for bandwidth. As for latency we haven't moved a bit.

True. These MT/s are good for bandwidth. (Sounds different if you skip the word "only".)

 

[zlobby]

True. These MT/s are good for bandwidth. (Sounds different if you skip the word "only".)

When the only two things RAM is useful for are latency and bandwidth (minus power draw), when latency is constant, then 'only' is rather implied. Dont'cha think?

 

[TranceHead]

Even outside how Windows looks at hard drives?

I don't fret when my hard drive displays less space, because i understand it's using 1024 × 1024.
Hard drives manufacturers use 1000 × 1000 to fluff the numbers up a bit.
In the old days, you had a 24MB HDD, it was seen as 24MB HDD because manufacturers were using the same base2 units DOS and Windows do.