# How fast can a PC be?

**URL:** <https://boards.straightdope.com/t/how-fast-can-a-pc-be/239825>\
**Category:** Factual Questions\
**Created:** [April 13, 2004, 6:14pm UTC](https://boards.straightdope.com/t/how-fast-can-a-pc-be/239825 "2004-04-13T18:14:20Z")\
**Posts on this page:** 11\
**Page:** 1

<div class="post-metadata">

**Author:** ![duffer](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/duffer/32/1026_2.png) [@duffer](https://boards.straightdope.com/u/duffer)\
**Post date:** [April 13, 2004, 6:14pm UTC](https://boards.straightdope.com/t/how-fast-can-a-pc-be/239825/1 "2004-04-13T18:14:20Z")

</div>

I know I’ve seen this answer somewhere, but can’t remember where. What I’m wondering is, what is the fastest speed a PC can eventually be before it plateaus? I don’t mean the Cray’s, but rather something that sits at your desk. Is there a theoretical limit?

---

<div class="post-metadata">

**Author:** ![Grave](https://avatars.discourse-cdn.com/v4/letter/g/71c47a/32.png) [@Grave](https://boards.straightdope.com/u/Grave)\
**Post date:** [April 13, 2004, 9:35pm UTC](https://boards.straightdope.com/t/how-fast-can-a-pc-be/239825/2 "2004-04-13T21:35:36Z")

</div>

The limit is based more on limits of current manufacturing processess. Moore’s Law might be what you are getting at - [decent writeup here.](http://www.arstechnica.com/paedia/m/moore/moore-1.html)

---

<div class="post-metadata">

**Author:** ![Headcoat](https://avatars.discourse-cdn.com/v4/letter/h/6bbea6/32.png) [@Headcoat](https://boards.straightdope.com/u/Headcoat)\
**Post date:** [April 13, 2004, 10:07pm UTC](https://boards.straightdope.com/t/how-fast-can-a-pc-be/239825/3 "2004-04-13T22:07:07Z")

</div>

The current limit is imposed by the speed of light and the inability to go faster than that. Eventually, there will be a plateau because the electrons will not be able to move any faster in the processor. Thats where quantum computers come in.

---

<div class="post-metadata">

**Author:** ![Quartz](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/quartz/32/267_2.png) [@Quartz](https://boards.straightdope.com/u/Quartz)\
**Post date:** [April 13, 2004, 10:14pm UTC](https://boards.straightdope.com/t/how-fast-can-a-pc-be/239825/4 "2004-04-13T22:14:20Z")

</div>

No cite, but I read somewhere that they’ve got a P4 up to 5 GHz with extreme cooling (liquid nitrogen IIRC). We’re already seeing a move to multiple execution units - be it CPUs, cores, or Hyperthreading or whatever. I see this continuoing, but I also see raw speed increasing.

---

<div class="post-metadata">

**Author:** ![One\_Brick\_Short\_of\_a\_Wall](https://avatars.discourse-cdn.com/v4/letter/o/8dc957/32.png) [@One\_Brick\_Short\_of\_a\_Wall](https://boards.straightdope.com/u/One_Brick_Short_of_a_Wall)\
**Post date:** [April 13, 2004, 10:26pm UTC](https://boards.straightdope.com/t/how-fast-can-a-pc-be/239825/5 "2004-04-13T22:26:48Z")

</div>

Check out this link to Tom’s Hardware where they overclocked a P4 to 5 ghz w/ liquid nitrogen: [http://www6.tomshardware.com/cpu/20031230/5ghz-13.html](http://www6.tomshardware.com/cpu/20031230/5ghz-13.html)

---

<div class="post-metadata">

**Author:** ![teemingONE](https://avatars.discourse-cdn.com/v4/letter/t/b4bc9f/32.png) [@teemingONE](https://boards.straightdope.com/u/teemingONE)\
**Post date:** [April 13, 2004, 10:31pm UTC](https://boards.straightdope.com/t/how-fast-can-a-pc-be/239825/6 "2004-04-13T22:31:23Z")

</div>

[hijack]

How fast would a computer have to be to mimic the speed in which information is sent in humans?

[/hijack]

---

<div class="post-metadata">

**Author:** ![Tuckerfan](https://avatars.discourse-cdn.com/v4/letter/t/eada6e/32.png) [@Tuckerfan](https://boards.straightdope.com/u/Tuckerfan)\
**Post date:** [April 14, 2004, 1:25am UTC](https://boards.straightdope.com/t/how-fast-can-a-pc-be/239825/7 "2004-04-14T01:25:30Z")

</div>

> [@teemingONE](#):
>
> [hijack]
> 
> How fast would a computer have to be to mimic the speed in which information is sent in humans?
> 
> [/hijack]

Well, it depends upon which bodily process you’re talking about. Some can take, I think, several minutes to get from one part of the body to the brain. There was an interview in _Wired_ (IIRC) with one of the guys from Sun who’d done a lot of experimentation with bandwidth transmissions, trying to see what the ultimate speed of the internet needed to be. He claimed that in the end they discovered that it didn’t really matter how fast your machine was, once the transmission rate exceeded something like 100 gigs a second, it didn’t matter how great the increase was, the person sitting at the machine couldn’t tell the difference. (So why they don’t increase the bus/port speeds of PCs to be capable of that, I’ve no idea.)

---

<div class="post-metadata">

**Author:** ![Max\_Torque](https://avatars.discourse-cdn.com/v4/letter/m/e9bcb4/32.png) [@Max\_Torque](https://boards.straightdope.com/u/Max_Torque)\
**Post date:** [April 14, 2004, 3:05am UTC](https://boards.straightdope.com/t/how-fast-can-a-pc-be/239825/8 "2004-04-14T03:05:34Z")

</div>

Mine can run an infinite loop in six seconds.

---

<div class="post-metadata">

**Author:** ![engineer\_comp\_geek](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/engineer_comp_geek/32/504_2.png) [@engineer\_comp\_geek](https://boards.straightdope.com/u/engineer_comp_geek)\
**Post date:** [April 14, 2004, 7:12am UTC](https://boards.straightdope.com/t/how-fast-can-a-pc-be/239825/9 "2004-04-14T07:12:16Z")

</div>

Back when I was in college, 1 MHz processors were the state of the art (ok, it was a long time ago). I very clearly remember a conversation with one of my college professors about how fast we would be able to make computers before we would start running into limitations of physics. He was quite sure that once we reached 40 MHz or so that the speeds would start to plateau.

Obviously, that didn’t happen.

Every year, you can have the exact same conversation, and pretty much every year you’ll come up with a number that’s a factor of 50 or so faster than the current best technology, but so far engineers have come up with creative ways to keep boosting the speeds at a pace that presently shows no signs of slowing down.

Whenever someone comes up with a number, I just think yep, we’ve had this conversation before. I’ll believe it when I see it.

FYI - the famous old Cray supercomputers, while impressive in their day, weren’t much faster than a high end desktop that you can buy at the local computer store. You can expect that the desktop PC in about 10 to 20 years will have close to the power of one of today’s supercomputers.

> [@teemingONE](#):
>
> [hijack]
> 
> How fast would a computer have to be to mimic the speed in which information is sent in humans?
> 
> [/hijack]

Computers are already significantly faster than the way information is sent through the human body. IIRC, humans have a “clock speed” that is about a thousand times slower than the typical desktop PC. The reason the human brain can do some things significantly faster than a computer is that the human brain benefits from massive parallelization, where a computer has to channel everything through one relatively tiny pipe.

There is no structure in the human brain that is capable of doing dedicated math functions like a computer, and there is also no structure in the human brain that is capable of moving large chunks of organized data from one place to another. If you need to add a thousand numbers very quickly, the computer will beat the brain every time. But, if you pick up an apple and hold it in your hand, your brain immediately tells you (within a fraction of a second) that what you have is in fact an apple, which it identifies out of the millions of different objects you have ever encountered in your life, and immediately associates taste, smell, and other information stored in your brain about all things apple-ish. A modern supercomputer, when faced with a similar task, can’t even come close.

---

<div class="post-metadata">

**Author:** ![jatfield](https://avatars.discourse-cdn.com/v4/letter/j/59ef9b/32.png) [@jatfield](https://boards.straightdope.com/u/jatfield)\
**Post date:** [April 14, 2004, 7:17am UTC](https://boards.straightdope.com/t/how-fast-can-a-pc-be/239825/10 "2004-04-14T07:17:28Z")

</div>

> [@Max Torque](#):
>
> Mine can run an infinite loop in six seconds.

ROFL, that’s great!

(hmm, probably my last post, and it’s spam. Drat!)

---

<div class="post-metadata">

**Author:** ![Padeye](https://avatars.discourse-cdn.com/v4/letter/p/a9a28c/32.png) [@Padeye](https://boards.straightdope.com/u/Padeye)\
**Post date:** [April 14, 2004, 1:47pm UTC](https://boards.straightdope.com/t/how-fast-can-a-pc-be/239825/11 "2004-04-14T13:47:12Z")

</div>

It’s a thankless task to esimate how fast computers can get. That is to say there is no limit but faster computers will have different technology than we use now so it’s hard to make estimates on what doesn’t exist yet. This is particularly true of overclocking. I would not have foreseen my now mediocre 1.15GHz Athlon CPU based on an estimate from the original 1MHz 6502 or 4.77MHz 8088. Heck, I once owned a Sanyo MBC-550, the only DOS PC I am aware of that was actually slower than an original IBM PC.
