iPhone 6 Is a Thin, Sexy Phone with a Killer Camera (application, blackberry)

Originally Posted by NJBest

Not everyone has a good understanding of technology. The push to 64 bit was evolutionary. Everyone was going to do it. But to be first to do it is purely marketing.

It's not like Apple created the ARM64 instruction set. It's not like they created the first ARM64 CPU either. Apple tries to one-up the competition, just like the competition tries to one-up them. ARM64 wasn't a revolutionary feature for Apple or anyone at this point. To speak of it as if it was, is cursory at best.

It's no more "purely marketing" than cramming in ever higher ppi displays. In fact it's far better, because it actually improves performance, as opposed to killing performance and battery life. Apple one upped the competition with something that's actually useful. That's quite a bit better than one upping the competition with silly gimmicks (following your eye movements, hover scrolling, anyone?) and pointless spec sheet bumps that benefit no one.

Who's claiming Apple created ARM64? And who's claiming it was revolutionary?

You do have one thing right though, a lot of people don't have a good understanding of technology, so they get sucked into stupid megapixel or ppi contests while dismissing the significance of being first to be putting 64 bit ARMv8 chips in the hands of consumers (not some FPGA demo). It's not just that it's 64 bit (nice future-proofing and some performance gains), it's also other improvements that result in things like massive, order of magnitude encryption speed increases.

If it had been some Android device that was first, the same people dismissing what Apple did would be slamming Apple for being behind.

[ericsami]

Originally Posted by JasonF

it's also other improvements that result in things massive, order of magnitude encryption speed increases.

you keep using this word and I am not sure you fully comprehend the weight of such words.

Let's say I have device X and it executes 1000 Instructions / MicroSecond

Then I have Device Y that executes 10,000 instructions / Microsecond.

Then we call Device Y is 1 order of magnitude faster than Device X.

This means it is roughly 10 times faster than the previous device.

I personally have not see any Microprocessor based technology speed to improve by ten folds in one year or even two years.

Originally Posted by ericsami

you keep using this word and I am not sure you fully comprehend the weight of such words.

Let's say I have device X and it executes 1000 Instructions / MicroSecond

Then I have Device Y that executes 10,000 instructions / Microsecond.

Then we call Device Y is 1 order of magnitude faster than Device X.

This means it is roughly 10 times faster than the previous device.

I personally have not see any Microprocessor based technology speed to improve by ten folds in one year or even two years.

I know exactly what it means. I posted the benchmarks. That you chose not to read them is your own issue. ARMv8 brought new cryptography instructions that resulted in 8.25x improvements for some tests. AKA about an order of magnitude.

[NJBest]

Originally Posted by JasonF

It's no more "purely marketing" than cramming in ever higher ppi displays. In fact it's far better, because it actually improves performance, as opposed to killing performance and battery life. Apple one upped the competition with something that's actually useful. That's quite a bit better than one upping the competition with silly gimmicks (following your eye movements, hover scrolling, anyone?) and pointless spec sheet bumps that benefit no one.

Who's claiming Apple created ARM64? And who's claiming it was revolutionary?

You do have one thing right though, a lot of people don't have a good understanding of technology, so they get sucked into stupid megapixel or ppi contests while dismissing the significance of being first to be putting 64 bit ARMv8 chips in the hands of consumers (not some FPGA demo). It's not just that it's 64 bit (nice future-proofing and some performance gains), it's also other improvements that result in things like massive, order of magnitude encryption speed increases.

If it had been some Android device that was first, the same people dismissing what Apple did would be slamming Apple for being behind.

That's an interesting anti-Apple approach you have there. Focus on performance than user experience. ARM64 has no real benefits today. Higher PPI screens do. They provide sharper images, and a better overall user experience. Yes, there's a cost to a better user experience, but there's a cost to larger registers as well.

Apple has hardly upped the competition significantly with ARM64 because it's not as useful as you incorrectly claim. It's just as useless as all those other things you mentioned. None of them are bad to try, however. Iterative design with quick feedback loops are fine.

I wasn't suggesting that anyone claimed that Apple created ARM64. I was educating you and others who are not familiar with technology.

Putting something that isn't really useful today into the hands of consumers isn't a feat. When they are able to use it in a meaningful manner, then we can talk. You should look into how encryption is used on devices. When was the last time you were waiting for an encryption callback hook? Even on older iPhones, when invoking obj C data handler... when has the callback not been ready?

Update socket io with a loop through, and send through data that requires a any read/write state.. even to a viewcontroller. Log the time you're waiting for the encryption callback hook. You'll see that there's no wait in any use case. If something is in a wait state for 0ms, then making it faster is hardly useful.

This is a modified version of socket data handler for iOS to demonstrate my point.

- (void) onDataNSString *)data
{
DEBUGLOG(@"onData %@", data);

[self setTimeout];

NSString *regex = @"^([^:]+)[0-9]+)?(\\+)?[^:]+)?:?(.*)?$";
NSString *regexPieces = @"^([0-9]+)(\\+)?(.*)";

NSMutableArray *test = [self getMatchesFrom:data with:regex];

if ([test count] > 0) {
NSArray *result = [test objectAtIndex:0];

int idx = [[result objectAtIndex:1] intValue];
SocketIOPacket *packet = [[SocketIOPacket alloc] initWithTypeIndex:idx];

packet.pId = [result objectAtIndex:2];

packet.ack = [result objectAtIndex:3];
packet.endpoint = [result objectAtIndex:4];
packet.data = [result objectAtIndex:5];

switch (idx) {
case 0: {
DEBUGLOG(@"disconnect");
[self onDisconnect:[NSError errorWithDomain:SocketIOError
code:SocketIOServerRespondedWithDisconnect
userInfo:nil]];
break;
}
case 1: {
DEBUGLOG(@"connected");
[self onConnectacket];
break;
}
case 2: {
DEBUGLOG(@"heartbeat");
[self sendHeartbeat];
break;
}
case 3: {
DEBUGLOG(@"message");
if (packet.data && ![packet.data isEqualToString:@""]) {
if ([_delegate respondsToSelector:@selector(socketIO:didReceiveMe ssage]) {
[_delegate socketIO:self didReceiveMessageacket];
}
}
break;
}
case 4: {
DEBUGLOG(@"json");
if (packet.data && ![packet.data isEqualToString:@""]) {
if ([_delegate respondsToSelector:@selector(socketIO:didReceiveJS ON]) {
[_delegate socketIO:self didReceiveJSONacket];
}
}
break;
}
case 5: {
DEBUGLOG(@"event");
if (packet.data && ![packet.data isEqualToString:@""]) {
NSDictionary *json = [packet dataAsJSON];
packet.name = [json objectForKey:@"name"];
packet.args = [json objectForKey:@"args"];
if ([_delegate respondsToSelector:@selector(socketIO:didReceiveEv ent]) {
[_delegate socketIO:self didReceiveEventacket];
}
}
break;
}
case 6: {
DEBUGLOG(@"ack");

NSMutableArray *pieces = [self getMatchesFromacket.data with:regexPieces];

if ([pieces count] > 0) {
NSArray *piece = [pieces objectAtIndex:0];
int ackId = [[piece objectAtIndex:1] intValue];
DEBUGLOG(@"ack id found: %d", ackId);

NSString *argsStr = [piece objectAtIndex:3];
id argsData = nil;
if (argsStr && ![argsStr isEqualToString:@""]) {
argsData = [SocketIOJSONSerialization objectFromJSONData:[argsStr dataUsingEncoding:NSUTF8StringEncoding] error:nil];
if (!_returnAllDataFromAck && [argsData count] > 0) {
argsData = [argsData objectAtIndex:0];
}
}

NSString *key = [NSString stringWithFormat:@"%d", ackId];
SocketIOCallback callbackFunction = [_acks objectForKey:key];
if (callbackFunction != nil) {
callbackFunction(argsData);
[self removeAcknowledgeForKey:key];
}
}

break;
}
case 7: {
DEBUGLOG(@"error");
break;
}
case 8: {
DEBUGLOG(@"noop");
break;
}
default: {
DEBUGLOG(@"command not found or not yet supported");
break;
}
}

packet = nil;
}
else {
DEBUGLOG(@"ERROR: data that has arrived wasn't valid");
}
}

[wolf39us]

Originally Posted by JasonF

Told by who? Did you actually try it?

I did, and what was true is no longer. They wouldn't ship to your house during pre-order stage. But they will after pre-orders.

I was just there today and managed to get one shipped to another person. But they said "that wasn't possible during preorder"

[NJBest]

Originally Posted by JasonF

You should probably call them up and tell them the huge volume of iPhones theyve arranged to ship to people's houses actually need to go to stores instead.

To what purpose? Do you think Verizon would actually listen? Have you actually done this yourself? Give it a try.

Originally Posted by NJBest

That's an interesting anti-Apple approach you have there. Focus on performance than user experience. ARM64 has no real benefits today. Higher PPI screens do. They provide sharper images, and a better overall user experience. Yes, there's a cost to a better user experience, but there's a cost to larger registers as well.

High PPI screens have a benefit up to a point. What some Android device manufacturers are doing serves no useful purpose and only matters when looking at a spec sheet. It hurts performance, hurts battery life, and provides nothing to the end user. You may reject physics but it's still true.

Apple has hardly upped the competition significantly with ARM64 because it's not as useful as you incorrectly claim. It's just as useless as all those other things you mentioned. None of them are bad to try, however. Iterative design with quick feedback loops are fine.

Really? Vastly improved crytography performance is useless? 10-20% across the board performance gains just from recompiling are useless?

I wasn't suggesting that anyone claimed that Apple created ARM64. I was educating you and others who are not familiar with technology.

I can assure I don't need an education, certainly not from you of all people. That's, quite frankly, laughable.

Now back to denying the improvements seen from a simple recompile...

Originally Posted by wolf39us

I did, and what was true is no longer. They wouldn't ship to your house during pre-order stage. But they will after pre-orders.

I was just there today and managed to get one shipped to another person. But they said "that wasn't possible during preorder"

Of course they would. I ordered two.

Actually I wound up ordering four - two online, then I changed my mind and ordered a different one over the phone, and the CS guy messed up and canceled both, so then I re-ordered it online.

[wolf39us]

Originally Posted by JasonF

Of course they would. I ordered two.

Actually I wound up ordering four - two online, then I changed my mind and ordered a different one over the phone, and the CS guy messed up and canceled both, so then I re-ordered it online.

dude, you're just not listening -- in my area, at MY stores they would NOT.

Originally Posted by wolf39us

dude, you're just not listening -- in my area, at MY stores they would NOT.

So we've gone from Verizon won't ship to your house, to Verizon won't ship to your house in your area, to the Verizon stores you went to wouldn't ship it to your house.

That's quite the walk back from your original blanket statement. Why would you go to a store to pre-order a phone that you want delivered to your house? And were they actual Verizon stores, owned by Verizon?