Start up company Tachyum claims to design a processor that can replace cpus and gpus

[Master Disaster]

1 minute ago, M.Yurizaki said:

So about this mystery processor some startup claims will revolutionize the hardware industry

It's about as realistic as a 2,800bhp car that can do 350mph.

[Mira Yurizaki]

Just now, Master Disaster said:

It's about as realistic as a 2,800bhp car that can do 350mph.

You forgot the 100MPG part.

[Drak3]

3 minutes ago, M.Yurizaki said:

So about this mystery processor some startup claims will revolutionize the hardware industry

If it turns out that what they claim is true, it's likely that the first dozen(s) of production attempts fail.

[Guest]

3 hours ago, gabrielcarvfer said:

Aren't both synapses eletrochemical? The electric synapses have the gap junction, but AFAIK still has ions moving across it, instead only propagating electrons.

A neuron only has one synapse at the bouton. There are electric synapses, but when talking about the brain we don't need to worry. Those are mainly involved in reflexes and can't be facilitated. We are mainly focusing on cholinergic, dopaminergic and GABA receptors, off the top of my head. Electric synapses would only occur in the spinal column, which reflexes are fully controlled by. 

[mynameisjuan]

1 hour ago, Drak3 said:

The point is that while possible,

How many people have to tell you not possible? Do you have this hidden secret that car companies would love to know? 

 

If that was possible at least one all gas car would have 100mpg but we are not even half way....

[Drak3]

4 minutes ago, mynameisjuan said:

How many people have to tell you not possible? Do you have this hidden secret that car companies would love to know? 

 

If that was possible at least one all gas car would have 100mpg but we are not even half way....

First, car companies are capable of making cars that achieve 100MPG. The issue is that they can't make cost effective vehicles that achieve that yet.

 

Second, stop adding fucking details that were never part of the original discussion.

 

The only criteria was:

1. Car
2. 100MPG
3. 350MPH

Which is possible, but not feasible.

 

Oh, by the way, the BMW i3 can achieve 124+ MPG.

 

So far, YOU are the only one to say, outright, that it is not possible. Everyone else said, to the effect, that it isn't happening with current technology and production cars won't achieve that.

[spartaman64]

5 minutes ago, Drak3 said:

First, car companies are capable of making cars that achieve 100MPG. The issue is that they can't make cost effective vehicles that achieve that yet.

 

Second, stop adding fucking details that were never part of the original discussion.

 

The only criteria was:

1. Car
2. 100MPG
3. 350MPH

Which is possible, but not feasible.

 

Oh, by the way, the BMW i3 can achieve 124+ MPG.

 

So far, YOU are the only one to say, outright, that it is not possible. Everyone else said, to the effect, that it isn't happening with current technology and production cars won't achieve that.

if you have an electric car that can go 350mph then you have a car with infinite MPG that can do 350mph lol

[mynameisjuan]

2 minutes ago, Drak3 said:

First, car companies are capable of making cars that achieve 100MPG. The issue is that they can't make cost effective vehicles that achieve that yet.

 

Second, stop adding fucking details that were never part of the original discussion.

 

The only criteria was:

1. Car
2. 100MPG
3. 350MPH

Which is possible, but not feasible.

 

Oh, by the way, the BMW i3 can achieve 124+ MPG.

 

So far, YOU are the only one to say, outright, that it is not possible. Everyone else said, to the effect, that it isn't happening with current technology and production cars won't achieve that.

 

Your post.

 

A car that can go 350mph and get 100mpg at lower speeds. Still not fucking possible. That was your comment, nothing to do with one car can do 350 and another can get 100mpg, not at all what people are saying. 

 

The i3 can push 100mpg in the city due to electric but cant get fucking close to 350. So yeah still falls claims like the OP.

[Drak3]

2 minutes ago, mynameisjuan said:

 

Your post.

 

A car that can go 350mph and get 100mpg at lower speeds. Still not fucking possible. That was your comment, nothing to do with one car can do 350 and another can get 100mpg, not at all what people are saying. 

 

The i3 can push 100mpg in the city due to electric but cant get fucking close to 350. So yeah still falls claims like the OP.

Reread EVERYTHING I've said CAREFULLY.

 

It is possible to do both with one vehicle, not at the same time, through a complex engine and powertrain system that, with current technology, would be prohibitively expensive and prohibitively unreliable.

 

I never said it would be a single engine system. I never said that it was an all gas system. I never said that it would be a production car. I never said it would happen with current production components. So stop putting fucking words in my mouth.

[Brooksie359]

2 hours ago, Drak3 said:

We aren't. The point is that while possible, with what we have today, it would be too expensive and hillariously unreliable to turn it into a product.

Key point is with what we have today. 

[Drak3]

1 minute ago, Brooksie359 said:

Key point is with what we have today. 

Again, with what we have today, it is possible, not feasible.

[suicidalfranco]

probably snake oil like Euclideon's unlimited detail (anyone remember them?)

[Mira Yurizaki]

2 hours ago, suicidalfranco said:

probably snake oil like Euclideon's unlimited detail (anyone remember them?)

What's funny is I think tech was also kind of old. 3DMark2001 had a benchmark called point sprites (it was one of the feature tests, with the horse) which seems suspiciously similar to what Unlimited Detail was trying to do.

[TheRandomness]

So basically what Larrabee could've been?
Haha yeah right.

[veli2501]

I feel like people are taking the car analogy far too literally.

 

 

More discussion about brain functions though please #interesting

[Dash Lambda]

On 6/26/2018 at 2:19 PM, Brooksie359 said:

That is under the assumption that gas is the only power source. Let's say you end up harnessing that wind resistance into a power source then the calculations would be meaningless. Again you are limiting what you think is possible based on what we currently have. The likely solution to something "impossible" would be something nobody has thought of yet. Real innovation doesn't come from talking about how it is impossible but rather asking how could it be possible. 

Cars actually already do use the wind resistance to improve performance. They add drag by channeling the air through radiators -Keeping things cool does more than just avoiding damage. As for harnessing that power, that's not how it works. The air isn't applying force to the car, the car is applying force to the air. In order to extract energy from the drag, you need to add drag -And you'll get less power out of it than you spend in overcoming that extra drag. It's like pulling yourself up by the bootstraps, you're pulling up but you're also pulling down.

So, while the drag can be utilized to assist power generation indirectly, using it to directly produce power wouldn't do anything useful.

 

We can never say that a solution to a problem is impossible, but we can say a specific solution or a solution meeting certain criteria is impossible. In this instance, propelling a given object through the atmosphere takes a certain amount of power, and you cannot exert that power without producing it.

[SpaceGhostC2C]

On 6/26/2018 at 7:26 AM, spartaman64 said:

I would think nvidia when designing a processor would want to be "reducing the number of slow wires on a chip, and reducing the average length of existing wires" so I doubt this was something that only they thought of. I'm skeptical of their claims of being superior to current processor companies like nvidia and intel but I guess only time will tell.

Your skepticism is healthy, and their quantitative claims are meaningless if they refer to products that don't exist yet.

 

But going to the specific bit I quoted, you have to consider that they are basically claiming incompatibility with everything. It's not a new CPU or GPU that you just run the same old benchmark on and obtain a better result. Instead, you have to code from scratch for it, or at the very least adapt somewhat so eventually you can compile it with their compiler. At that point, the comparison with CPUs or GPUs becomes irrelevant, since they're really going against Nvidia's / Intel's custom chips, or at best Xeon Phis (although the whole point of Phi was to still be x86-compatible, as opposed to GPUs, so it's not really even going against Phi).